{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HJ7MAJKSSCE7CJDWAVXUTMYC2F","short_pith_number":"pith:HJ7MAJKS","schema_version":"1.0","canonical_sha256":"3a7ec025529089f12476056f49b302d16a5ca2e5b9c6fba596311e2996429c43","source":{"kind":"arxiv","id":"2507.17448","version":1},"attestation_state":"computed","paper":{"title":"Reasoning-Driven Retrosynthesis Prediction with Large Language Models via Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","physics.chem-ph"],"primary_cat":"cs.CE","authors_text":"Bo Chen, Hanqi Li, Kai Yu, Lu Chen, Situo Zhang, Xin Chen, Xuanze Lin, Zichen Zhu, Zihan Zhao","submitted_at":"2025-07-23T12:13:06Z","abstract_excerpt":"Retrosynthesis planning, essential in organic synthesis and drug discovery, has greatly benefited from recent AI-driven advancements. Nevertheless, existing methods frequently face limitations in both applicability and explainability. Traditional graph-based and sequence-to-sequence models often lack generalized chemical knowledge, leading to predictions that are neither consistently accurate nor easily explainable. To address these challenges, we introduce RetroDFM-R, a reasoning-based large language model (LLM) designed specifically for chemical retrosynthesis. Leveraging large-scale reinfor"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2507.17448","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CE","submitted_at":"2025-07-23T12:13:06Z","cross_cats_sorted":["cs.AI","physics.chem-ph"],"title_canon_sha256":"9ed7432dd80675929e1c24dbcfddcfb71d7e8c024cf5eba675985177d23cde16","abstract_canon_sha256":"c76adfdb5b8ef0cfb48aba5d79a0733ebece180850ad666063e863b417a30bf3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:07.892517Z","signature_b64":"w6vbpZT7FJfbDjnTToyl0EzzCkEJddJ0fyy8Ju9e+BvpZGuzndQD3Enb+Fd2B1fApJDJfGh3CScrm/LBlXXcAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a7ec025529089f12476056f49b302d16a5ca2e5b9c6fba596311e2996429c43","last_reissued_at":"2026-07-05T11:42:07.892033Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:07.892033Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reasoning-Driven Retrosynthesis Prediction with Large Language Models via Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","physics.chem-ph"],"primary_cat":"cs.CE","authors_text":"Bo Chen, Hanqi Li, Kai Yu, Lu Chen, Situo Zhang, Xin Chen, Xuanze Lin, Zichen Zhu, Zihan Zhao","submitted_at":"2025-07-23T12:13:06Z","abstract_excerpt":"Retrosynthesis planning, essential in organic synthesis and drug discovery, has greatly benefited from recent AI-driven advancements. Nevertheless, existing methods frequently face limitations in both applicability and explainability. Traditional graph-based and sequence-to-sequence models often lack generalized chemical knowledge, leading to predictions that are neither consistently accurate nor easily explainable. To address these challenges, we introduce RetroDFM-R, a reasoning-based large language model (LLM) designed specifically for chemical retrosynthesis. Leveraging large-scale reinfor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.17448","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2507.17448/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2507.17448","created_at":"2026-07-05T11:42:07.892094+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.17448v1","created_at":"2026-07-05T11:42:07.892094+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.17448","created_at":"2026-07-05T11:42:07.892094+00:00"},{"alias_kind":"pith_short_12","alias_value":"HJ7MAJKSSCE7","created_at":"2026-07-05T11:42:07.892094+00:00"},{"alias_kind":"pith_short_16","alias_value":"HJ7MAJKSSCE7CJDW","created_at":"2026-07-05T11:42:07.892094+00:00"},{"alias_kind":"pith_short_8","alias_value":"HJ7MAJKS","created_at":"2026-07-05T11:42:07.892094+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11589","citing_title":"Large Language Models in Process Systems Engineering: Opportunities, Architectures, and Industrial Deployment Challenges","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12113","citing_title":"Augmenting Molecular Language Models with Local $n$-gram Memory","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07181","citing_title":"RETROSPECT: RETROsynthesis via Sequential Prediction, and Chemically Transformed-ranking","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F","json":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F.json","graph_json":"https://pith.science/api/pith-number/HJ7MAJKSSCE7CJDWAVXUTMYC2F/graph.json","events_json":"https://pith.science/api/pith-number/HJ7MAJKSSCE7CJDWAVXUTMYC2F/events.json","paper":"https://pith.science/paper/HJ7MAJKS"},"agent_actions":{"view_html":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F","download_json":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F.json","view_paper":"https://pith.science/paper/HJ7MAJKS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.17448&json=true","fetch_graph":"https://pith.science/api/pith-number/HJ7MAJKSSCE7CJDWAVXUTMYC2F/graph.json","fetch_events":"https://pith.science/api/pith-number/HJ7MAJKSSCE7CJDWAVXUTMYC2F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F/action/storage_attestation","attest_author":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F/action/author_attestation","sign_citation":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F/action/citation_signature","submit_replication":"https://pith.science/pith/HJ7MAJKSSCE7CJDWAVXUTMYC2F/action/replication_record"}},"created_at":"2026-07-05T11:42:07.892094+00:00","updated_at":"2026-07-05T11:42:07.892094+00:00"}