{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NGEB7FQLCJELYHBEDVGBIKHL6S","short_pith_number":"pith:NGEB7FQL","schema_version":"1.0","canonical_sha256":"69881f960b1248bc1c241d4c1428ebf48812d2790a8fb3590846046355a048f3","source":{"kind":"arxiv","id":"2410.01044","version":2},"attestation_state":"computed","paper":{"title":"RATIONALYST: Mining Implicit Rationales for Process Supervision of Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Andrew Wang, Benjamin Van Durme, Chuyu Liu, Daniel Khashabi, Dongwei Jiang, Guoxuan Wang, Jingyu Zhang, Yining Lu","submitted_at":"2024-10-01T20:05:51Z","abstract_excerpt":"The reasoning steps generated by LLMs might be incomplete, as they mimic logical leaps common in everyday communication found in their pre-training data: underlying rationales are frequently left implicit (unstated). To address this challenge, we introduce RATIONALYST, a model for process-supervision of reasoning based on pre-training on a vast collection of rationale annotations extracted from unlabeled data. We extract 79k rationales from web-scale unlabelled dataset (the Pile) and a combination of reasoning datasets with minimal human intervention. This web-scale pre-training for reasoning "},"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":"2410.01044","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-01T20:05:51Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"43a563f1fd65299d5d2009f6d671f61a2a988c54a08614f98d03df7a4ac6b3d1","abstract_canon_sha256":"d6afea56e3d60e209fa54693d0a8a417a09f92967a9273a5fcd5fd6ed9b20be8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:36.303874Z","signature_b64":"oywOgpsG0Tepi66W9ofTtQuFOieBx3HcBUZpWl6Nvs+mXA7vxWbkqzml0WsWTrh2p3UeaVETqQIBNz3IseTaDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"69881f960b1248bc1c241d4c1428ebf48812d2790a8fb3590846046355a048f3","last_reissued_at":"2026-07-05T11:21:36.303356Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:36.303356Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RATIONALYST: Mining Implicit Rationales for Process Supervision of Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Andrew Wang, Benjamin Van Durme, Chuyu Liu, Daniel Khashabi, Dongwei Jiang, Guoxuan Wang, Jingyu Zhang, Yining Lu","submitted_at":"2024-10-01T20:05:51Z","abstract_excerpt":"The reasoning steps generated by LLMs might be incomplete, as they mimic logical leaps common in everyday communication found in their pre-training data: underlying rationales are frequently left implicit (unstated). To address this challenge, we introduce RATIONALYST, a model for process-supervision of reasoning based on pre-training on a vast collection of rationale annotations extracted from unlabeled data. We extract 79k rationales from web-scale unlabelled dataset (the Pile) and a combination of reasoning datasets with minimal human intervention. This web-scale pre-training for reasoning "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01044","kind":"arxiv","version":2},"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/2410.01044/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":"2410.01044","created_at":"2026-07-05T11:21:36.303425+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.01044v2","created_at":"2026-07-05T11:21:36.303425+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01044","created_at":"2026-07-05T11:21:36.303425+00:00"},{"alias_kind":"pith_short_12","alias_value":"NGEB7FQLCJEL","created_at":"2026-07-05T11:21:36.303425+00:00"},{"alias_kind":"pith_short_16","alias_value":"NGEB7FQLCJELYHBE","created_at":"2026-07-05T11:21:36.303425+00:00"},{"alias_kind":"pith_short_8","alias_value":"NGEB7FQL","created_at":"2026-07-05T11:21:36.303425+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/NGEB7FQLCJELYHBEDVGBIKHL6S","json":"https://pith.science/pith/NGEB7FQLCJELYHBEDVGBIKHL6S.json","graph_json":"https://pith.science/api/pith-number/NGEB7FQLCJELYHBEDVGBIKHL6S/graph.json","events_json":"https://pith.science/api/pith-number/NGEB7FQLCJELYHBEDVGBIKHL6S/events.json","paper":"https://pith.science/paper/NGEB7FQL"},"agent_actions":{"view_html":"https://pith.science/pith/NGEB7FQLCJELYHBEDVGBIKHL6S","download_json":"https://pith.science/pith/NGEB7FQLCJELYHBEDVGBIKHL6S.json","view_paper":"https://pith.science/paper/NGEB7FQL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.01044&json=true","fetch_graph":"https://pith.science/api/pith-number/NGEB7FQLCJELYHBEDVGBIKHL6S/graph.json","fetch_events":"https://pith.science/api/pith-number/NGEB7FQLCJELYHBEDVGBIKHL6S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NGEB7FQLCJELYHBEDVGBIKHL6S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NGEB7FQLCJELYHBEDVGBIKHL6S/action/storage_attestation","attest_author":"https://pith.science/pith/NGEB7FQLCJELYHBEDVGBIKHL6S/action/author_attestation","sign_citation":"https://pith.science/pith/NGEB7FQLCJELYHBEDVGBIKHL6S/action/citation_signature","submit_replication":"https://pith.science/pith/NGEB7FQLCJELYHBEDVGBIKHL6S/action/replication_record"}},"created_at":"2026-07-05T11:21:36.303425+00:00","updated_at":"2026-07-05T11:21:36.303425+00:00"}