{"as_of":"2026-08-10T11:15:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:5148229f357e45d82e8858d9edcb459557da7d136c6550cc3cbd7d723aec0d33","coverage":[{"denominator":93,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":93,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-05T19:07:59.804892Z","state":"measured"},{"denominator":95,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":95,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-10T06:31:04.303077+00:00","state":"measured"},{"denominator":2,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":2,"source":"paper_references, paper_reference_links","source_observed_at":"2026-06-27T09:57:12.398344Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-07-03T10:37:56.687533Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"cited_work":{"arxiv_id":"2508.13408","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2508.13408","snapshot_observed_at":"2026-07-03T10:37:56.687533Z","title":"arXiv preprint arXiv:2508.13408 , year=","venue":null,"work_id":"8f544dfb-6888-4c07-a451-45dd9bf06595","year":null},"citing_paper":{"arxiv_id":"2605.10230","last_updated":"2026-05-11T09:06:42Z","snapshot_observed_at":"2026-08-09T16:26:16.483365Z","submitted_at":"2026-05-11T09:06:42Z","title":"FORGE: Fragment-Oriented Ranking and Generation for Context-Aware Molecular Optimization","version":1},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-05-12T05:25:48.965001Z"},"links":{"cited_paper":"/paper/2508.13408","citing_paper":"/paper/2605.10230"},"observation_digest":"sha256:b07907723e5ed2d4c2f4a3912b897e6bec79421999f534256f6b2d0d22bbdfe9","observation_id":"d09399db-bba3-4b79-9711-b7644b4857dd","resolution":{"observed_at":"2026-05-12T05:26:23.603249Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"cited_work":{"arxiv_id":"2508.13408","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2508.13408","snapshot_observed_at":"2026-07-03T10:37:56.687533Z","title":"arXiv preprint arXiv:2508.13408 , year=","venue":null,"work_id":"8f544dfb-6888-4c07-a451-45dd9bf06595","year":null},"citing_paper":{"arxiv_id":"2606.12113","last_updated":"2026-06-10T14:07:18Z","snapshot_observed_at":"2026-07-06T23:51:02.805629Z","submitted_at":"2026-06-10T14:07:18Z","title":"Augmenting Molecular Language Models with Local $n$-gram Memory","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-06-27T09:57:12.398344Z"},"links":{"cited_paper":"/paper/2508.13408","citing_paper":"/paper/2606.12113"},"observation_digest":"sha256:25ced3aca6c20cff5b3da029f2b05cb301ca7eaa4c099654ab13836cbb8776c7","observation_id":"51b8db38-d6dc-46ae-88a1-b0c2030b7b81","resolution":{"observed_at":"2026-07-03T10:37:56.689010Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2508.13408/citation-record","integrity":"/paper/2508.13408/integrity","json":"/paper/2508.13408/citation-record.json","paper":"/paper/2508.13408"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:47.409096Z","title":"Generative Pre - Training from Molecules , September 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:47.409096Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:e809c6358c88feff0312f973f4f531aa423934d0aaf3ff9e63966879664bb484","observation_id":"8875ca5e-1886-48f4-978e-66b9a4be1162","resolution":{"observed_at":"2026-08-05T19:07:47.409096Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:47.686279Z","title":"Fast, accurate, and reliable molecular docking with QuickVina 2","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:47.686279Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:9e28fb5c0bba70419ebee19c89252dad9682cda343456bdfe9d050b11d7979ec","observation_id":"ff37e80d-ebcf-4553-801f-9d79491ab86a","resolution":{"observed_at":"2026-08-05T19:07:47.686279Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:47.851306Z","title":null,"venue":null,"work_id":null,"year":2064},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:47.851306Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:5e44e770d722d37ba15d35dfb8a9645f02ed82f8176ea051e002165d1d49ae8f","observation_id":"1a1471db-7dd0-4ff6-be90-4088f710b6fc","resolution":{"observed_at":"2026-08-05T19:07:47.851306Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:47.973415Z","title":"Bemis and Mark A","venue":null,"work_id":null,"year":1996},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:47.973415Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:27dbeb8cb08efc3196fdfda461f0e97b0d418fdd2b342604b80bae1887c31546","observation_id":"fed5d498-0bc4-4a6b-92ba-950fdceae696","resolution":{"observed_at":"2026-08-05T19:07:47.973415Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:48.113493Z","title":"Acegen: Reinforcement learning of generative chemical agents for drug discovery","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:48.113493Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:314946bac7c0792b3a57e47abbbbcde144cb9be2473c887179db08477e65e043","observation_id":"cf8110cd-3a7d-4db5-a3f7-a7eb09d78ccc","resolution":{"observed_at":"2026-08-05T19:07:48.113493Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:48.271646Z","title":"GNN - FiLM : Graph Neural Networks with Feature -wise Linear Modulation","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:48.271646Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:084b02714cd586af887c3d9eed708616f13df2cedeeb67b1724adac8ca6bd55c","observation_id":"f12dcb8c-9aba-4c00-832a-c26313017337","resolution":{"observed_at":"2026-08-05T19:07:48.271646Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1805.11973","last_updated":"2022-09-27T10:04:29Z","snapshot_observed_at":"2026-08-07T21:51:58.025112Z","submitted_at":"2018-05-30T13:56:06Z","title":"MolGAN: An implicit generative model for small molecular graphs","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1805.11973","snapshot_observed_at":"2026-08-05T19:07:48.449111Z","title":"MolGAN : An implicit generative model for small molecular graphs, September 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:48.449111Z"},"links":{"cited_paper":"/paper/1805.11973","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:0c001acb9367c18f0de5226cf034ae0189ee340cd7adc67f0fe8b551aa6fd2a4","observation_id":"44db3cd4-57de-4f7d-98c5-e99d58205e62","resolution":{"observed_at":"2026-08-05T19:07:48.449111Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2211.16349","last_updated":"2022-11-29T16:30:53Z","snapshot_observed_at":"2026-08-06T19:48:32.253623Z","submitted_at":"2022-11-29T16:30:53Z","title":"BARTSmiles: Generative Masked Language Models for Molecular Representations","version":1},"cited_work":{"arxiv_id":"2211.16349","doi":null,"metadata_source":"pith","pith_arxiv_id":"2211.16349","snapshot_observed_at":"2026-08-05T19:08:10.834746Z","title":"BARTSmiles: Generative Masked Language Models for Molecular Representations","venue":"cs.LG","work_id":"e2a13f74-5dad-4e50-8a24-4d03ef70f160","year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:48.658271Z"},"links":{"cited_paper":"/paper/2211.16349","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:bd792ae54dafafaca73a4a76b5e6df48c9fa60a4666562bc32284a59be5b66c7","observation_id":"313ccc06-a907-4926-92db-d3d9572d30be","resolution":{"observed_at":"2026-08-05T19:08:10.954751Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:48.796822Z","title":"Fu, Stefano Ermon, Atri Rudra, and Christopher Ré","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:48.796822Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:a7482f31b76e76b0458a682a805cbfa1525788a5a2aebc258e0c9fb92d3e8158","observation_id":"1cb10d5c-a584-416b-8ba1-46a4f31b052e","resolution":{"observed_at":"2026-08-05T19:07:48.796822Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:48.970655Z","title":"On the Art of Compiling and Using ' Drug - Like ' Chemical Fragment Spaces","venue":null,"work_id":null,"year":2008},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:48.970655Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:936ad86aae8096677b85910dc317ccc8845913a5696453e0b058d3600a2c0b02","observation_id":"b646976c-9ad1-4470-b075-6b3762ccbd9e","resolution":{"observed_at":"2026-08-05T19:07:48.970655Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.21783","last_updated":"2024-11-23T23:27:33Z","snapshot_observed_at":"2026-07-06T18:55:11.576666Z","submitted_at":"2024-07-31T17:54:27Z","title":"The Llama 3 Herd of Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.21783","snapshot_observed_at":"2026-08-05T19:07:49.239142Z","title":"The llama 3 herd of models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:49.239142Z"},"links":{"cited_paper":"/paper/2407.21783","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:9797e048a26c396d3911d4a637c1902dae349f3116fc43939455528911cbe246","observation_id":"988d0570-3340-46b6-b9ce-5a3891ef325f","resolution":{"observed_at":"2026-08-05T19:07:49.239142Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:49.359629Z","title":"Durant, Burton A","venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:49.359629Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:f9ef92869ded8adf04365336b176dfdf54e2f34ff54065641c2886d2706f8d49","observation_id":"06141fc8-71c1-47d7-956a-5907daf78691","resolution":{"observed_at":"2026-08-05T19:07:49.359629Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:49.534681Z","title":"LIMO : Latent Inceptionism for Targeted Molecule Generation","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:49.534681Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:22184a092e4e73e1b968b594c89f5f9f502078533a191afc0810e21f8e88820a","observation_id":"43196b5c-3cc7-4157-9568-0a4ae0749c43","resolution":{"observed_at":"2026-08-05T19:07:49.534681Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:49.680665Z","title":"Bioreason: Incentivizing multimodal biological reasoning within a dna-llm model","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:49.680665Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:413a43876607e6153fdc5ead9f425c0f0f88debc0b3af4b23bfac9cc101df6f8","observation_id":"fd16d9fe-1efa-40fa-93ed-8d1792d42032","resolution":{"observed_at":"2026-08-05T19:07:49.680665Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:49.787225Z","title":"Geometry-enhanced molecular representation learning for property prediction","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:49.787225Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:f9df0447c64d098033ff58b2926968670e230a8c3cd1a7a9a4d4df7f7a1c6511","observation_id":"1183a795-cecd-4db4-bec2-75f221ca5410","resolution":{"observed_at":"2026-08-05T19:07:49.787225Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:49.897072Z","title":"Domain- Agnostic Molecular Generation with Chemical Feedback","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:49.897072Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:dbc272415c57ca780016d8326624859372b2d8554e3e7d7bc2955929001d5169","observation_id":"e89fed57-3c46-4bb6-a738-559357ce6ae4","resolution":{"observed_at":"2026-08-05T19:07:49.897072Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:50.116672Z","title":"Neural scaling of deep chemical models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:50.116672Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:e144cc666a724ae2500fc69faf5828907ef6e114a0c830cf66aadb60ace15a0a","observation_id":"a83da620-0377-44c0-8c8c-792bd0d202c5","resolution":{"observed_at":"2026-08-05T19:07:50.116672Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2206.12411","last_updated":"2022-10-09T18:15:38Z","snapshot_observed_at":"2026-07-06T13:24:27.607054Z","submitted_at":"2022-06-22T20:36:49Z","title":"Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2206.12411","snapshot_observed_at":"2026-08-05T19:07:50.235756Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:50.235756Z"},"links":{"cited_paper":"/paper/2206.12411","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:22d45831f052f5d13a0a7b867126be7f35680b3db38a7ba070262be55825fab4","observation_id":"0fa081b4-0a1d-46ef-81cb-48deb4f5a831","resolution":{"observed_at":"2026-08-05T19:07:50.235756Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:50.334742Z","title":null,"venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:50.334742Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:e53523929b91ba852e28e36ba8fce6b97011cd6fbf7a3c3c1b69d822b0b7fde9","observation_id":"c18881a2-6397-48e1-ac47-0d0509dd70a9","resolution":{"observed_at":"2026-08-05T19:07:50.334742Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:50.472399Z","title":"Schoenholz, Patrick F","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:50.472399Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:b38b98d4f416f2a42e06ad21a34de6315634fc1130ef53f77c21cfc44b2cd9e4","observation_id":"8991cbf0-1843-4d5c-a177-727baf2eda43","resolution":{"observed_at":"2026-08-05T19:07:50.472399Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1021/acs.jcim.9b00943","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"Bidirectional Molecule Generation with Recurrent Neural Networks","venue":"CyberLeninK (CyberLeninka)","work_id":"51669f8f-43d5-49bc-9480-6b9463a67445","year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:50.547538Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:755de6ce579515daa9f982def5a349f0a39bee82cd0315b3f951aa1f834e70de","observation_id":"e1096ef8-6118-43bd-a27b-3d69ef081f65","resolution":{"observed_at":"2026-08-05T19:08:04.986582Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2402.00838","last_updated":"2024-06-07T21:59:52Z","snapshot_observed_at":"2026-07-06T17:23:51.547578Z","submitted_at":"2024-02-01T18:28:55Z","title":"OLMo: Accelerating the Science of Language Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.00838","snapshot_observed_at":"2026-08-05T19:07:50.614454Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:50.614454Z"},"links":{"cited_paper":"/paper/2402.00838","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:c27b34d907c3e38ab62989341cd839456026c8ba6488d7ffb9b1154f9003aae4","observation_id":"913b39a8-0dc8-4e49-b568-ecd9152b8315","resolution":{"observed_at":"2026-08-05T19:07:50.614454Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2405.17066","last_updated":"2024-05-27T11:37:36Z","snapshot_observed_at":"2026-08-06T15:32:09.308687Z","submitted_at":"2024-05-27T11:37:36Z","title":"Saturn: Sample-efficient Generative Molecular Design using Memory Manipulation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.17066","snapshot_observed_at":"2026-08-05T19:07:50.816649Z","title":"Saturn: Sample-efficient generative molecular design using memory manipulation","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:50.816649Z"},"links":{"cited_paper":"/paper/2405.17066","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:7f272d244f76e2584577e8fcef767460920245e9d7c6997228087e0a1773be4b","observation_id":"37b60ffc-222e-4480-9d6d-b989c27c9af9","resolution":{"observed_at":"2026-08-05T19:07:50.816649Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:50.997545Z","title":"Scaffold splits overestimate virtual screening performance","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:50.997545Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:e10ba4407a60e112cfc601b7a6819431645fb485824c656a79e4626658a4bf67","observation_id":"10016855-c50b-4b27-90dc-039743739973","resolution":{"observed_at":"2026-08-05T19:07:50.997545Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.24963/ijcai.2023/744","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"Iyer, Yihong Ma, Olaf Wiest, Xiangliang Zhang, Wei Wang, Chuxu Zhang, and Nitesh V","venue":null,"work_id":"7246668e-f3ef-40fd-8849-cee3caf4a176","year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:51.091277Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:be72c6a51c66055ee234927abc5f0b807228e3d594af13ee580e25fc75576e14","observation_id":"c8efa27a-743e-4dc9-a3c1-6b59fbaa5269","resolution":{"observed_at":"2026-08-05T19:08:04.604751Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:51.194771Z","title":"Wei, David Duvenaud, José Miguel Hernández-Lobato, Benjamín Sánchez-Lengeling, Dennis Sheberla, Jorge Aguilera-Iparraguirre, Timothy D","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:51.194771Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:eb530ea5a49c6db43a4cb086fbdd2bab78ef22ae7a013fe613c08526633df36a","observation_id":"835d0f3a-f597-4ba3-ba76-91731ab92bc0","resolution":{"observed_at":"2026-08-05T19:07:51.194771Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:51.334828Z","title":"Denoising diffusion probabilistic models","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:51.334828Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:91d12ee391e4eb0b6896e2ccf18566ebc51b7405b3e1c13f19bc2527dfefe274","observation_id":"1fb27c38-1b24-4cf5-ad83-591fa707c644","resolution":{"observed_at":"2026-08-05T19:07:51.334828Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:51.524789Z","title":"Equivariant Diffusion for Molecule Generation in 3D","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:51.524789Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:4f53234f606d9f6ed11cac9b1badd50bd32c793465a28276071d3f0e7f5a0ffc","observation_id":"5364ea7a-15d7-44de-9701-1e3371cfebd2","resolution":{"observed_at":"2026-08-05T19:07:51.524789Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:51.624763Z","title":"MDM : Molecular Diffusion Model for 3D Molecule Generation","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:51.624763Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:0d0793eae05494bf44b0e5a47a5094c8bdd345b5bda9d94570930326a070299c","observation_id":"d3823e88-3f93-4067-b125-cb456799704f","resolution":{"observed_at":"2026-08-05T19:07:51.624763Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:51.814831Z","title":"Zinc: a free tool to discover chemistry for biology","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:51.814831Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:42ff9e8126602934c3221cc549372fd0de180d2d51f9c84aae739c29c12e9367","observation_id":"52527e07-aa84-4685-a1f6-d01b7e65bc90","resolution":{"observed_at":"2026-08-05T19:07:51.814831Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:51.994748Z","title":"Chemformer: a pre-trained transformer for computational chemistry","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:51.994748Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:63f56d6a5ea4060f5436d03134f0228a1dd84dcae7f0bbb5c65604d33612a730","observation_id":"1a280d74-3cd5-4b44-9ac1-85a270d5b28b","resolution":{"observed_at":"2026-08-05T19:07:51.994748Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:52.159909Z","title":"Autonomous molecule generation using reinforcement learning and docking to develop potential novel inhibitors","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:52.159909Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:e1e59dc6a9ef425551ed9bf36e5c2c4b623a8249ff7229d327e6b92a63a7d5fd","observation_id":"bb63907c-444b-4ee6-ab45-aa78c2db8791","resolution":{"observed_at":"2026-08-05T19:07:52.159909Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.06825","last_updated":"2023-10-10T17:54:58Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-10-10T17:54:58Z","title":"Mistral 7B","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.06825","snapshot_observed_at":"2026-08-05T19:07:52.324654Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:52.324654Z"},"links":{"cited_paper":"/paper/2310.06825","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:f9f957ecd86964015e7356ff9770b7e798ff17083a3e92893d7277c911e3668b","observation_id":"e6d81fad-805a-4559-b6fe-a39e6e7839f4","resolution":{"observed_at":"2026-08-05T19:07:52.324654Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:19.196850Z","title":"Junction Tree Variational Autoencoder for Molecular Graph Generation","venue":null,"work_id":"ec3b656e-e461-4db1-9145-70a8ea0ac975","year":2018},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:52.534764Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:386ec06642c203c7f9e0eee0d6fbfe80f287bf9f7c81813416e3489c8ee926a2","observation_id":"9f1036be-a4ed-4ae4-83bc-4829ec19a65f","resolution":{"observed_at":"2026-08-05T19:08:19.404759Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:18.914757Z","title":"Multi- Objective Molecule Generation using Interpretable Substructures","venue":null,"work_id":"117e5a44-6296-4903-ac62-bc9f75d2dd9b","year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:52.794754Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:99090eadec4b40b679e4e6ccfbcd73581e9c9e92aee189c081e86957b955ca67","observation_id":"dfe64f6b-9562-435f-9a9e-ae81e7a27037","resolution":{"observed_at":"2026-08-05T19:08:19.024838Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:18.574752Z","title":"Hierarchical generation of molecular graphs using structural motifs","venue":null,"work_id":"2249b07b-1773-447a-b42f-2871a92672a2","year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:52.936015Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:fdad2ea4550ba9654295f4745927d02440cb6e19b7a4fd1ddb38b7b074a9bf20","observation_id":"a401930b-fc98-448e-9634-1d8daac1a95b","resolution":{"observed_at":"2026-08-05T19:08:18.724755Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:18.259443Z","title":"Score-based Generative Modeling of Graphs via the System of Stochastic Differential Equations","venue":null,"work_id":"2446532f-69db-44aa-aee9-7f7580f9986c","year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:53.080511Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:08b9d2236a9493206ff346914ad2c4d0d0b0ae5e635aed7d2091f1cad6b43070","observation_id":"bb3602de-e80e-4035-b876-8de5da7cf05f","resolution":{"observed_at":"2026-08-05T19:08:18.394749Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:17.924749Z","title":"Kipf and Max Welling","venue":null,"work_id":"507d2a55-6964-4f7c-9ae2-8853cbc6a937","year":2017},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":38,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:53.186970Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:d9f18073d9b366b43560b032ca5df4cc012db1888d2139f48a8f9ab44bb0550b","observation_id":"7ab61c8e-8cf6-44bd-b8a4-7354cc462c3f","resolution":{"observed_at":"2026-08-05T19:08:18.094833Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1038/432823a","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"Chemical space","venue":"Nature","work_id":"c8f64e6d-89e1-4f48-8aff-99e0563aae73","year":2004},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:53.398029Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:fb3d70623df68af25ad64413ce902545afd6acaaca9c3f1042587dd99dc95e54","observation_id":"9ef3cb62-5a5e-4685-92d2-363723d5e3d5","resolution":{"observed_at":"2026-08-05T19:08:03.814755Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:17.464756Z","title":"Aditya Prakash, and Chao Zhang","venue":null,"work_id":"4e8796ba-65f6-47bc-abdf-cf1894dbfb6c","year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:53.664749Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:e9f725d40be90519ea3be129a0cbf00be56d01718b1ad07b7f523de5b8980473","observation_id":"af7c6c35-b03d-4c48-8a04-ee145a755796","resolution":{"observed_at":"2026-08-05T19:08:17.654752Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2022.10058","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:09.245578Z","title":"Frey, Pascal Friederich, Théophile Gaudin, Alberto Alexander Gayle, Kevin Maik Jablonka, Rafael F","venue":null,"work_id":"47a13bc8-a76b-4626-a54c-456207c43e88","year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:53.861805Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:b7b26a955785e3369eaa17a85827f5a0377ec00128ce2525d49be93c4c64b214","observation_id":"13ff30c6-084c-4484-b360-1d51b9f4c62d","resolution":{"observed_at":"2026-08-05T19:08:09.484751Z","resolver_source":"raw_fallback","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:54.038726Z","title":"Subword Regularization : Improving Neural Network Translation Models with Multiple Subword Candidates","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:54.038726Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:6f24ee61e3efa6689b9da8cb7b1cefb9c05f6f664c76f4cfae25a77f8d403885","observation_id":"c55b65db-5177-43bb-9e9c-72ad80881afd","resolution":{"observed_at":"2026-08-05T19:07:54.038726Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1609/aaai.v35i9.17001","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"MolGrow : A Graph Normalizing Flow for Hierarchical Molecular Generation","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","work_id":"c94f6a7d-b80c-41a5-8d8c-21db1ccb1c61","year":2021},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:54.204750Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:974f0c0a9e31ed8aeeb85db93688793c81bd6a55b9471f1856a94f1c5bce83be","observation_id":"a514ee22-df93-4d56-b16a-c05444ccf48a","resolution":{"observed_at":"2026-08-05T19:08:03.394761Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1039/d0sc03115a","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"Neuraldecipher – reverse-engineering extended-connectivity fingerprints ( ECFPs ) to their molecular structures","venue":"Chemical Science","work_id":"d265c814-e099-4e63-9edf-a14892d206c0","year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":44,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:54.314751Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:d9a1c9c432e979de155f42977972355ce1a0bc02ae8943d0e6b6d8e2902c3f8e","observation_id":"4722ab2a-fd69-4c0c-8042-a2e2fdd184fa","resolution":{"observed_at":"2026-08-05T19:08:02.935296Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:17.131747Z","title":"Exploring Chemical Space with Score -based Out -of-distribution Generation","venue":null,"work_id":"580a82fc-5387-43f4-9e81-7edf6e1f139c","year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":45,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:54.430115Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:83947620990cba8e3874b19600a05a56fada4bbf59705685a5ee021503cae681","observation_id":"e7031da3-8c4a-4913-8936-cceae3b4e7a5","resolution":{"observed_at":"2026-08-05T19:08:17.274753Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:16.785545Z","title":"Molecule Generation with Fragment Retrieval Augmentation","venue":null,"work_id":"da529a71-ab16-4ee3-9e8a-2780b325c677","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:54.557004Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:dc834dc620f7ca647178a0b3ac0624f0ce28de32c572412a112a16f73417db75","observation_id":"e8e635bd-422b-4b5c-b9cc-386c2907db1a","resolution":{"observed_at":"2026-08-05T19:08:16.909709Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:16.444760Z","title":"Drug discovery with dynamic goal-aware fragments","venue":null,"work_id":"9e9ffb38-1f9d-4b1c-890a-8eb57b56b627","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":47,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:54.694751Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:99bf248eb85d71428eee0feb29c428a503cf809f2ea28910103f5d9a613e023b","observation_id":"242a39aa-ddc7-43ec-af44-305da64d3836","resolution":{"observed_at":"2026-08-05T19:08:16.594884Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:16.009659Z","title":"Limits to depth-efficiencies of self-attention","venue":null,"work_id":"e776b2e1-5881-4bbf-99d8-ea1443fbc697","year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":48,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:54.870851Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:f66eb86db1713680fb2890eac16faf445ed50323116d7167373a7de418ac8fc1","observation_id":"765262cf-5621-4129-a148-cae3ae1896b9","resolution":{"observed_at":"2026-08-05T19:08:16.194753Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1093/bioinformatics/btae558","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"Chemical reaction enhanced graph learning for molecule representation","venue":"Bioinformatics","work_id":"255a3825-6e14-435c-adba-83824acafbbe","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":49,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.068211Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:536566fcb9e6a052b9af0216d2a2cee39fc6b85481afa7be73f287f2a5f301ac","observation_id":"4101b0fe-cedb-4d97-b5db-639d58ccb17b","resolution":{"observed_at":"2026-08-05T19:08:02.549929Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1609/aaai.v36i4.20377","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"GeomGCL : Geometric Graph Contrastive Learning for Molecular Property Prediction","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","work_id":"a7faa99c-bc3b-4871-b143-a1cbfe4f70da","year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":50,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.293557Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:5c5b75fdfd6cc3c66f0a92512de6b1f35938a67442cc688f0bbbbeacf4cf649f","observation_id":"07d418f5-0b3a-41a5-bc00-1f53e70141b1","resolution":{"observed_at":"2026-08-05T19:08:02.295931Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:15.694750Z","title":"A 3D generative model for structure-based drug design","venue":null,"work_id":"042137ad-9f73-4328-a07a-85b934206c9b","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":51,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.394751Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:9c49c31413202f8490b372db1a6ed0c793f6356ef5976b5f13ff634a541584f4","observation_id":"c2026c92-6d45-4a1b-8f09-bf090541d381","resolution":{"observed_at":"2026-08-05T19:08:15.844753Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1038/s41467-021-23415-2","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"Masked graph modeling for molecule generation","venue":"Nature Communications","work_id":"3249774b-ffbb-4638-b4dd-324f49a16562","year":2021},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":52,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.499796Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:9dfe5eca21e34d4542d4f2097a76c1e897f56764fcdc0f90c597eb38195c765d","observation_id":"2444502f-fd5c-4df7-9877-c5ecd4301ed4","resolution":{"observed_at":"2026-08-05T19:08:01.954922Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:15.334878Z","title":"Provably Powerful Graph Networks","venue":null,"work_id":"1a5f8037-92b3-4979-b688-18d6d8482ba6","year":2019},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.532810Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:13380d8d5ceb557733918d32e00e1ebe809849cd56258b71bf86f578c7c580e7","observation_id":"13ce5048-e3e4-4a7a-9857-a71434881728","resolution":{"observed_at":"2026-08-05T19:08:15.464760Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1038/s41598-023-35648-w","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"Molecule generation using transformers and policy gradient reinforcement learning","venue":"Scientific Reports","work_id":"9e0e33fa-09ad-4ac5-b3c4-0f08f14f7371","year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":54,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.607139Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:6d92848864a1e814edd89e26fcddc25a6cbc4b2d52d0e8870eca46f887ecbc82","observation_id":"e1517adc-6704-4388-84b5-93343cb8f394","resolution":{"observed_at":"2026-08-05T19:08:01.594013Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:55.704864Z","title":"Hamilton, Jan Eric Lenssen, Gaurav Rattan, and Martin Grohe","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":55,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.704864Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:92528a618d17d948849a48a71920a22e04778cbe823da4abdc35252f2cb6b372","observation_id":"a74a20b6-06ee-4318-bcb9-cc3b88fd4c93","resolution":{"observed_at":"2026-08-05T19:07:55.704864Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:55.814954Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":56,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.814954Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:71a7bf350605baf8352482d68a59be24da4c46f7d94f366c82fbcbc2821c1057","observation_id":"982159c3-81ce-47a6-b2f3-2d8e0d708458","resolution":{"observed_at":"2026-08-05T19:07:55.814954Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:14.978424Z","title":"DeepSMILES : An Adaptation of SMILES for Use in Machine - Learning of Chemical Structures , September 2018","venue":null,"work_id":"33d30ab8-f6a2-4260-b182-843db3084c90","year":2018},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":57,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:55.898183Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:1d735b85df95d8dae234accaf39e93530e014f829aafdbc1d7e8ed4e3ffae3f4","observation_id":"98afe171-6ba3-410f-90f5-ee0cb3aca5b1","resolution":{"observed_at":"2026-08-05T19:08:15.154755Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:56.014777Z","title":"Molecular de-novo design through deep reinforcement learning","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":58,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:56.014777Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:ab92f00fee37927b344546461af4e6b42db556bc85f8b475bc827d4bb9e4886c","observation_id":"3d28ff52-310a-400a-8064-1a4d24d9dd3e","resolution":{"observed_at":"2026-08-05T19:07:56.014777Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2501.05457","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:08.584904Z","title":"The jungle of generative drug discovery: Traps, treasures, and ways out","venue":null,"work_id":"f25642f6-4d9c-4276-ac40-f862c2bf83ed","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":59,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:56.124759Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:e64d0fdde0012e1dae8b089a914fc7f61571de7a72ed07ab8b6fe167448510d7","observation_id":"877afc25-cc9f-41d2-b563-fc4f1652afdc","resolution":{"observed_at":"2026-08-05T19:08:08.854853Z","resolver_source":"raw_fallback","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:14.668678Z","title":"A Deep Generative Model for Fragment - Based Molecule Generation","venue":null,"work_id":"07c80fe4-ba42-4f60-ab9f-3709f9fa4a30","year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":60,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:56.244758Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:1a17bfd8449a4c45523c71e5d387750edbb38f9af02a9908bbeeb991a9d56aa4","observation_id":"65a37bab-f849-4169-83dd-494fd7c64e64","resolution":{"observed_at":"2026-08-05T19:08:14.768670Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:56.304998Z","title":"Molecular Sets ( MOSES ): A Benchmarking Platform for Molecular Generation Models","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":61,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:56.304998Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:109b83932aa1f626bb6eb63f70355206e3fce14790ad347f1ec28b596f053cbd","observation_id":"f8c754ee-8359-4013-adf8-055d1b6194c5","resolution":{"observed_at":"2026-08-05T19:07:56.304998Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:56.394852Z","title":"Extended- Connectivity Fingerprints","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":62,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:56.394852Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:79f18a3e949be9a323908dcfdc4e338d8bea88df93bd71952380a2d52f39c8fa","observation_id":"077155cc-b3c5-4d63-9e85-aa7854eaf088","resolution":{"observed_at":"2026-08-05T19:07:56.394852Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2106.09553","last_updated":"2022-12-14T14:52:16Z","snapshot_observed_at":"2026-07-06T11:20:23.028632Z","submitted_at":"2021-06-17T14:33:55Z","title":"Large-Scale Chemical Language Representations Capture Molecular Structure and Properties","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2106.09553","snapshot_observed_at":"2026-08-05T19:07:56.624854Z","title":"Large- Scale Chemical Language Representations Capture Molecular Structure and Properties , December 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":63,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:56.624854Z"},"links":{"cited_paper":"/paper/2106.09553","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:1711a6f1d41aeac27779fbbeb2a58a3f8001e98a38b88dca87765f8df9221cff","observation_id":"5f66fa65-32be-4139-8ef5-05c646f529f2","resolution":{"observed_at":"2026-08-05T19:07:56.624854Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2405.04912","last_updated":"2025-03-31T18:16:41Z","snapshot_observed_at":"2026-07-06T18:11:26.210335Z","submitted_at":"2024-04-04T16:20:06Z","title":"GP-MoLFormer: A Foundation Model For Molecular Generation","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.04912","snapshot_observed_at":"2026-08-05T19:07:56.714877Z","title":"Hoffman, Vijil Chenthamarakshan, Youssef Mroueh, and Payel Das","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":64,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:56.714877Z"},"links":{"cited_paper":"/paper/2405.04912","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:b5fac93e456d9e78516280a8b998f8441a6abd1867fb2a64b27b23391bc12616","observation_id":"c90d3af8-993c-4c05-8111-613bea310c6e","resolution":{"observed_at":"2026-08-05T19:07:56.714877Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:56.855317Z","title":"Blum, and Jean-Louis Reymond","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":65,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:56.855317Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:3a0d7ae26092606d43771cb87aa19c81de6512d392db825db623fca03d658fdf","observation_id":"a3d9670a-ec01-4e8f-8f7b-052f73563e7e","resolution":{"observed_at":"2026-08-05T19:07:56.855317Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1707.06347","last_updated":"2017-08-28T09:20:06Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2017-07-20T02:32:33Z","title":"Proximal Policy Optimization Algorithms","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1707.06347","snapshot_observed_at":"2026-08-05T19:07:57.030884Z","title":"Proximal policy optimization algorithms","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":66,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:57.030884Z"},"links":{"cited_paper":"/paper/1707.06347","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:c67ec73a7d4f629d3fe4eb1e856c6b8779960ae342b7a6bb5594afcd27cd0178","observation_id":"a05fa6bd-74d0-4bc7-852d-2ffa30cc296f","resolution":{"observed_at":"2026-08-05T19:07:57.030884Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:57.158862Z","title":"Hunter, Costas Bekas, and Alpha A","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":67,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:57.158862Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:a06825d52968338dc2749804f084872b8e5a27f01619f9103e2c52555a003293","observation_id":"e55c7225-69bb-4aee-bc02-7914a0bcc3de","resolution":{"observed_at":"2026-08-05T19:07:57.158862Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:57.259423Z","title":"Neural Machine Translation of Rare Words with Subword Units","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":68,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:57.259423Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:81276d2fcc149151ee53e16a83865c223a6f5af6415647158b11ebe26a2113c0","observation_id":"8ecc8eb0-5973-4c12-ba47-4e6f24dd62e3","resolution":{"observed_at":"2026-08-05T19:07:57.259423Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1802.03480","last_updated":"2018-02-09T23:57:46Z","snapshot_observed_at":"2026-08-01T16:06:18.416569Z","submitted_at":"2018-02-09T23:57:46Z","title":"GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1802.03480","snapshot_observed_at":"2026-08-05T19:07:57.363912Z","title":"GraphVAE : Towards Generation of Small Graphs Using Variational Autoencoders , February 2018","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":69,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:57.363912Z"},"links":{"cited_paper":"/paper/1802.03480","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:553affba4ad4a82b0e82b013f97340f15bc1335c5eef280ba61cf5be27821f2e","observation_id":"9617f998-1731-40c3-b2fe-495640092940","resolution":{"observed_at":"2026-08-05T19:07:57.363912Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:14.334755Z","title":"Chemical language models enable navigation in sparsely populated chemical space","venue":null,"work_id":"b3e7a486-6456-4b5a-b694-f99bba1123dd","year":2021},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":70,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:57.444889Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:590929f926296c041fb7031172f80171eaf98378b6434354c2a0a4ae50dedf77","observation_id":"ae282a47-c6ba-41e0-95cc-e988350a6d1c","resolution":{"observed_at":"2026-08-05T19:08:14.465945Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:14.074753Z","title":"Deep Unsupervised Learning using Nonequilibrium Thermodynamics","venue":null,"work_id":"ae0e698a-e0d0-4c9d-9c13-f770f210c1a4","year":2015},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":71,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:57.556788Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:28e8b7b9919861cd02fb989673b0716fee9a43513b3783af7f5c15a1656d8130","observation_id":"832ef6b8-2a2a-4b5b-b7d3-f4bc37ba483c","resolution":{"observed_at":"2026-08-05T19:08:14.194789Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2211.09086","last_updated":"2022-11-16T18:07:43Z","snapshot_observed_at":"2026-08-07T18:47:56.988002Z","submitted_at":"2022-11-16T18:07:43Z","title":"Molecular Fingerprints for Robust and Efficient ML-Driven Molecular Generation","version":1},"cited_work":{"arxiv_id":"2211.09086","doi":null,"metadata_source":"pith","pith_arxiv_id":"2211.09086","snapshot_observed_at":"2026-08-05T19:08:07.215693Z","title":"Molecular Fingerprints for Robust and Efficient ML-Driven Molecular Generation","venue":"cs.LG","work_id":"4b078a6a-eccc-482f-955c-1f09d020f055","year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":72,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:57.786978Z"},"links":{"cited_paper":"/paper/2211.09086","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:1b7a6a6052b2877f85d292aa130ff266c1ce309dab8cbcfe6b7b56df0926e785","observation_id":"41cf5932-a777-4859-a55a-621866ee9e82","resolution":{"observed_at":"2026-08-05T19:08:07.324739Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:13.726039Z","title":"Augmented hill-climb increases reinforcement learning efficiency for language-based de novo molecule generation","venue":null,"work_id":"05cc79ba-52e5-43fd-9c11-f5ae167456eb","year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":73,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:57.891306Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:6c78d1bf14b8157a9a80d7eeed47131e594b59492507048a29a6a84c9e61edc8","observation_id":"868cfe40-5196-4d6b-9b44-6f2e21bcd2c6","resolution":{"observed_at":"2026-08-05T19:08:13.844851Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:13.265331Z","title":"Promptsmiles: prompting for scaffold decoration and fragment linking in chemical language models","venue":null,"work_id":"f609cf69-76af-4155-a1ab-7f65bc125756","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":74,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.004584Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:d84e0a748ed0a62b12cb319f081f9f93e38f32942fd015318bfa4ef1eb87885d","observation_id":"cfe71c45-32ae-4b0d-ba4b-f07329c6e9d6","resolution":{"observed_at":"2026-08-05T19:08:13.355296Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:58.146912Z","title":"Tingle, Khanh G","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":75,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.146912Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:9e226a8309974936abd3913eaa7e88bcf3d69ba969423a936b0961d8ec1f074c","observation_id":"7586eed3-0446-440f-b1fa-b14eb8a6c9bd","resolution":{"observed_at":"2026-08-05T19:07:58.146912Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.09267","last_updated":"2023-10-13T17:25:11Z","snapshot_observed_at":"2026-08-09T08:16:38.647148Z","submitted_at":"2023-10-13T17:25:11Z","title":"Genetic algorithms are strong baselines for molecule generation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.09267","snapshot_observed_at":"2026-08-05T19:07:58.221964Z","title":"Genetic algorithms are strong baselines for molecule generation","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":76,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.221964Z"},"links":{"cited_paper":"/paper/2310.09267","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:9f54b43f3041fa11e9cfd208476dd7865545860e08f748d3529ce5680020f77e","observation_id":"aed08a07-8c39-4f49-b40d-9116533decc3","resolution":{"observed_at":"2026-08-05T19:07:58.221964Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:58.280640Z","title":"cMolGPT : A Conditional Generative Pre - Trained Transformer for Target - Specific De Novo Molecular Generation","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":77,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.280640Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:6fcf3da18db1a1085d7e2d56a7e1c6251d70b61b32d6a1616e7cf46533f7f449","observation_id":"4168959b-fff1-4600-a894-2b7ee9f4a9db","resolution":{"observed_at":"2026-08-05T19:07:58.280640Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:58.354763Z","title":"SMILES , a chemical language and information system","venue":null,"work_id":null,"year":1988},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":78,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.354763Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:8813499056ab31211f6615cd7f15a3529c739eb8585dd615bde90772fb5b6a7a","observation_id":"d7f91b31-4e75-42c5-96d6-8447c65ac69f","resolution":{"observed_at":"2026-08-05T19:07:58.354763Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1039/c9sc01928f","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T05:30:23.456663Z","title":"Efficient multi-objective molecular optimization in a continuous latent space","venue":"Chemical Science","work_id":"87e4a7b1-bb25-46e9-8ae0-83537495b788","year":2019},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":79,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.479390Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:8b3a2cfa7ccf2a9eb6852def13826257929d06dadeda8c35ddc69e26da3fc640","observation_id":"0e1a3150-4f44-4168-8aeb-4e0920bf7387","resolution":{"observed_at":"2026-08-05T19:08:00.405559Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1910.03771","last_updated":"2020-07-14T03:42:34Z","snapshot_observed_at":"2026-07-06T08:27:58.343233Z","submitted_at":"2019-10-09T03:23:22Z","title":"HuggingFace's Transformers: State-of-the-art Natural Language Processing","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1910.03771","snapshot_observed_at":"2026-08-05T19:07:58.584900Z","title":"Huggingface's transformers: State-of-the-art natural language processing","venue":null,"work_id":null,"year":1910},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":80,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.584900Z"},"links":{"cited_paper":"/paper/1910.03771","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:1a103c4b442a2db2c18d06f0b2c689b0eae88a447d6996f5bae2ffd1bc05bbfb","observation_id":"a9926c63-daa2-42d8-9d60-520d4197e3ca","resolution":{"observed_at":"2026-08-05T19:07:58.584900Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:12.948526Z","title":"MoleculeNet: a benchmark for molecular machine learning","venue":null,"work_id":"b7d33d06-38e3-41c2-8a11-558bf0bcdf1c","year":2018},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":81,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.694767Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:5d13f02d2ba9a7450bbf3078649dd5efafb81ea9008f075e2fbb8ff0b4cade5e","observation_id":"829cb8a2-2902-4cab-a76f-aed4249045b5","resolution":{"observed_at":"2026-08-05T19:08:13.074921Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:58.745532Z","title":"Graph neural networks for automated de novo drug design","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":82,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.745532Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:e7728a92a157bdff71843ca438e01ae11284236cce630a25a042db9b6dafef8f","observation_id":"eb3a09b7-2752-429b-a62a-28d901d33767","resolution":{"observed_at":"2026-08-05T19:07:58.745532Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:58.854753Z","title":"Pushing the Boundaries of Molecular Representation for Drug Discovery with the Graph Attention Mechanism","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":83,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.854753Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:7679070e484101f826332bd64668c0eb96ba7464a1055cd3f1f223778f50e85d","observation_id":"4554f454-584b-4276-a24f-120bd2b0d958","resolution":{"observed_at":"2026-08-05T19:07:58.854753Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:12.644742Z","title":"How Powerful are Graph Neural Networks ? In International Conference on Learning Representations, September 2018","venue":null,"work_id":"c93b40ac-3ce0-414d-b5bd-e253b150f285","year":2018},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":84,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:58.935328Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:ebfcb468c27806a424e88c32e1f74b668136ad55ccedb1333aa7a362c2f0e21c","observation_id":"9371aef9-b43f-4cd6-abe8-7864d6653f60","resolution":{"observed_at":"2026-08-05T19:08:12.734740Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:12.342008Z","title":"Powers, Ron O","venue":null,"work_id":"d17cc679-d97e-49d9-89ea-1765f7de9680","year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":85,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.051041Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:6a8fe8c455b472cbd04330b9efa686d89ec618ed631d6ea5f074690a06df950e","observation_id":"ccc9c278-da74-4759-9d13-fc5a12f79835","resolution":{"observed_at":"2026-08-05T19:08:12.425243Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:59.153448Z","title":"Hit and lead discovery with explorative rl and fragment-based molecule generation","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":86,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.153448Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:63792e9438546e36c2294d3041f1bac4f0ecb66aabadfb13c450ef616b4f12ec","observation_id":"1b1ffb58-8c0b-4874-afa4-83107d501bbb","resolution":{"observed_at":"2026-08-05T19:07:59.153448Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:11.904751Z","title":"Hit and lead discovery with explorative RL and fragment-based molecule generation","venue":null,"work_id":"6505a4ef-2f3f-4782-a844-5755eb576e69","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":87,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.237076Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:a4b153d6eb1ee691ebf26319fc7a3caf61c19bc68dad31bd1663d4d08786c5f2","observation_id":"feed1bb7-5b03-4250-8208-552a2fd486d0","resolution":{"observed_at":"2026-08-05T19:08:12.084752Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:11.625145Z","title":"Llasmol: Advancing large language models for chemistry with a large-scale, comprehensive, high-quality instruction tuning dataset","venue":null,"work_id":"bb60ee07-b8f5-4968-b262-29bd82d71620","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":88,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.306984Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:fc0b225fce298372d3df75fa1b2ab24df87fc2fc0ba9fc8887914d1f1e495627","observation_id":"078b14f4-0d01-45cd-992c-36cfff34f0e7","resolution":{"observed_at":"2026-08-05T19:08:11.733938Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:59.384751Z","title":"MoFlow : An Invertible Flow Model for Generating Molecular Graphs","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":89,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.384751Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:abeddbab35241057a20fb7105725dd918415c98df01e41bb987ada548ffbd1b7","observation_id":"a8e26dc6-a4bd-4294-875c-6c67182ecafc","resolution":{"observed_at":"2026-08-05T19:07:59.384751Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:59.496086Z","title":null,"venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":90,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.496086Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:18893ecdc90f28802532f178cd875afaf83359b2467db9d85b07df22a450bf1c","observation_id":"84536375-c48b-4094-a638-4a994439fb18","resolution":{"observed_at":"2026-08-05T19:07:59.496086Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:07:59.574745Z","title":"ResGen is a pocket-aware 3D molecular generation model based on parallel multiscale modelling","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":91,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.574745Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:33cf3491bedd087965876ef9e86ca170355d493e433bd5ff1716aee89dd71fbb","observation_id":"64915037-b021-4200-b2b8-8bf1821c664b","resolution":{"observed_at":"2026-08-05T19:07:59.574745Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.03686","last_updated":"2024-06-06T02:10:50Z","snapshot_observed_at":"2026-07-06T18:26:15.151211Z","submitted_at":"2024-06-06T02:10:50Z","title":"BindGPT: A Scalable Framework for 3D Molecular Design via Language Modeling and Reinforcement Learning","version":1},"cited_work":{"arxiv_id":"2406.03686","doi":null,"metadata_source":"pith","pith_arxiv_id":"2406.03686","snapshot_observed_at":"2026-08-05T19:08:05.894751Z","title":"BindGPT: A Scalable Framework for 3D Molecular Design via Language Modeling and Reinforcement Learning","venue":"cs.LG","work_id":"e6b63984-2f2c-4d89-8a14-654136bd8582","year":2024},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":92,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.704753Z"},"links":{"cited_paper":"/paper/2406.03686","citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:45412cfc9f7e4957b41b826c5e6384585fb00dc72275655b15e8a5525d94da7c","observation_id":"4fc907c1-b453-487c-8048-6aa096cd2229","resolution":{"observed_at":"2026-08-05T19:08:06.075160Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T19:08:11.344748Z","title":"Uni- Mol : A Universal 3D Molecular Representation Learning Framework","venue":null,"work_id":"569932fb-02c6-442f-9557-74164eced296","year":2022},"citing_paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining","version":2},"reference_index":93,"source":"arxiv_source","source_observed_at":"2026-08-05T19:07:59.804892Z"},"links":{"citing_paper":"/paper/2508.13408"},"observation_digest":"sha256:d642d6e218e1573217a948747c7c2c1770c7917eddb0f1ab7184199290ff5eda","observation_id":"a8336139-c6f4-4bc9-88c3-1e22f7de96aa","resolution":{"observed_at":"2026-08-05T19:08:11.454746Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2508.13408","last_updated":"2025-08-22T13:02:19Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-06T19:48:25.824651Z","submitted_at":"2025-08-19T00:04:48Z","title":"NovoMolGen: Rethinking Molecular Language Model Pretraining"},"reference_resolution":{"displayed":93,"state_counts":{"malformed_identifier":0,"metadata_mismatch":2,"parse_uncertain":0,"unresolved":54,"verified_exact":13,"verified_fuzzy":24},"total_outbound_references":93},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"thesis":"As of 10 August 2026, this Paper Citation Record lists 93 of 93 outbound references and 2 inbound Pith citation observations for arXiv:2508.13408."}