{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:6HEDCBJCJVNPM5O7S626YZGT3X","short_pith_number":"pith:6HEDCBJC","schema_version":"1.0","canonical_sha256":"f1c83105224d5af675df97b5ec64d3ddd64b929aa06d36d0426d8d1abfce8e6d","source":{"kind":"arxiv","id":"2301.13755","version":3},"attestation_state":"computed","paper":{"title":"Retrosynthetic Planning with Dual Value Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Austin Tripp, Di Xue, Guoqing Liu, Krzysztof Maziarz, Marwin Segler, Shufang Xie, Tao Qin, Tie-Yan Liu, Yingce Xia, Zongzhang Zhang","submitted_at":"2023-01-31T16:43:53Z","abstract_excerpt":"Retrosynthesis, which aims to find a route to synthesize a target molecule from commercially available starting materials, is a critical task in drug discovery and materials design. Recently, the combination of ML-based single-step reaction predictors with multi-step planners has led to promising results. However, the single-step predictors are mostly trained offline to optimize the single-step accuracy, without considering complete routes. Here, we leverage reinforcement learning (RL) to improve the single-step predictor, by using a tree-shaped MDP to optimize complete routes. Specifically, w"},"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":"2301.13755","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-01-31T16:43:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"338382bd87ca4671e291903ed5b6aefbb411c01d513c94ffb28a04f4716c3eb2","abstract_canon_sha256":"dcceb6933c38d748c5fd742eab0ae53d7e04a8b4b9dac08ec958f55376e4b331"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:51:26.205836Z","signature_b64":"sV9GpDNtKmu5XOCR3agNd6lvBIB2nNd130dsqyQ01AUTVxjSKKoa2I4kzks3clqMB3CBgBor8Xf32wRJa2iCCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f1c83105224d5af675df97b5ec64d3ddd64b929aa06d36d0426d8d1abfce8e6d","last_reissued_at":"2026-07-05T07:51:26.205405Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:51:26.205405Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Retrosynthetic Planning with Dual Value Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Austin Tripp, Di Xue, Guoqing Liu, Krzysztof Maziarz, Marwin Segler, Shufang Xie, Tao Qin, Tie-Yan Liu, Yingce Xia, Zongzhang Zhang","submitted_at":"2023-01-31T16:43:53Z","abstract_excerpt":"Retrosynthesis, which aims to find a route to synthesize a target molecule from commercially available starting materials, is a critical task in drug discovery and materials design. Recently, the combination of ML-based single-step reaction predictors with multi-step planners has led to promising results. However, the single-step predictors are mostly trained offline to optimize the single-step accuracy, without considering complete routes. Here, we leverage reinforcement learning (RL) to improve the single-step predictor, by using a tree-shaped MDP to optimize complete routes. Specifically, w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.13755","kind":"arxiv","version":3},"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/2301.13755/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":"2301.13755","created_at":"2026-07-05T07:51:26.205464+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.13755v3","created_at":"2026-07-05T07:51:26.205464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.13755","created_at":"2026-07-05T07:51:26.205464+00:00"},{"alias_kind":"pith_short_12","alias_value":"6HEDCBJCJVNP","created_at":"2026-07-05T07:51:26.205464+00:00"},{"alias_kind":"pith_short_16","alias_value":"6HEDCBJCJVNPM5O7","created_at":"2026-07-05T07:51:26.205464+00:00"},{"alias_kind":"pith_short_8","alias_value":"6HEDCBJC","created_at":"2026-07-05T07:51:26.205464+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X","json":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X.json","graph_json":"https://pith.science/api/pith-number/6HEDCBJCJVNPM5O7S626YZGT3X/graph.json","events_json":"https://pith.science/api/pith-number/6HEDCBJCJVNPM5O7S626YZGT3X/events.json","paper":"https://pith.science/paper/6HEDCBJC"},"agent_actions":{"view_html":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X","download_json":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X.json","view_paper":"https://pith.science/paper/6HEDCBJC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.13755&json=true","fetch_graph":"https://pith.science/api/pith-number/6HEDCBJCJVNPM5O7S626YZGT3X/graph.json","fetch_events":"https://pith.science/api/pith-number/6HEDCBJCJVNPM5O7S626YZGT3X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X/action/storage_attestation","attest_author":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X/action/author_attestation","sign_citation":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X/action/citation_signature","submit_replication":"https://pith.science/pith/6HEDCBJCJVNPM5O7S626YZGT3X/action/replication_record"}},"created_at":"2026-07-05T07:51:26.205464+00:00","updated_at":"2026-07-05T07:51:26.205464+00:00"}