{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ZSECQMXWOASIOESPBFO6JQLVIB","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"3d3972f906d78cfd62f3984698a1a4326edcdf10c6af2744edd1c7e80d98d4dc","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T15:02:56Z","title_canon_sha256":"81ce3e176326d5c2742d6facd25a0701da7de5d6358c1cd98c9d28da7e49593c"},"schema_version":"1.0","source":{"id":"2402.06457","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.06457","created_at":"2026-07-05T08:55:07Z"},{"alias_kind":"arxiv_version","alias_value":"2402.06457v2","created_at":"2026-07-05T08:55:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.06457","created_at":"2026-07-05T08:55:07Z"},{"alias_kind":"pith_short_12","alias_value":"ZSECQMXWOASI","created_at":"2026-07-05T08:55:07Z"},{"alias_kind":"pith_short_16","alias_value":"ZSECQMXWOASIOESP","created_at":"2026-07-05T08:55:07Z"},{"alias_kind":"pith_short_8","alias_value":"ZSECQMXW","created_at":"2026-07-05T08:55:07Z"}],"graph_snapshots":[{"event_id":"sha256:b70d061ea8963e5dadc166ae51ff6f53fa60da7dca22afd65a31d44d8335f67c","target":"graph","created_at":"2026-07-05T08:55:07Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2402.06457/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Common self-improvement approaches for large language models (LLMs), such as STaR, iteratively fine-tune LLMs on self-generated solutions to improve their problem-solving ability. However, these approaches discard the large amounts of incorrect solutions generated during this process, potentially neglecting valuable information in such solutions. To address this shortcoming, we propose V-STaR that utilizes both the correct and incorrect solutions generated during the self-improvement process to train a verifier using DPO that judges correctness of model-generated solutions. This verifier is us","authors_text":"Aaron Courville, Alessandro Sordoni, Arian Hosseini, Nikolay Malkin, Rishabh Agarwal, Xingdi Yuan","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T15:02:56Z","title":"V-STaR: Training Verifiers for Self-Taught Reasoners"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.06457","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:42601e32c2e90aa05c126d5034266a8bc7ff0608778bdd0ccb2a158b1067e52b","target":"record","created_at":"2026-07-05T08:55:07Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"3d3972f906d78cfd62f3984698a1a4326edcdf10c6af2744edd1c7e80d98d4dc","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T15:02:56Z","title_canon_sha256":"81ce3e176326d5c2742d6facd25a0701da7de5d6358c1cd98c9d28da7e49593c"},"schema_version":"1.0","source":{"id":"2402.06457","kind":"arxiv","version":2}},"canonical_sha256":"cc882832f6702487124f095de4c175404bf1444485084e35163444d0a50cc6a6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cc882832f6702487124f095de4c175404bf1444485084e35163444d0a50cc6a6","first_computed_at":"2026-07-05T08:55:07.884739Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:55:07.884739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/kcF8AYmbfgdmAJ6V2CpLKBwnw+VvYao40/jOl/1RMlBzTIDqZFc7yn06dN8jtK5xhWKJzEJjRlP8PL+fE4VCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:55:07.885226Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.06457","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:42601e32c2e90aa05c126d5034266a8bc7ff0608778bdd0ccb2a158b1067e52b","sha256:b70d061ea8963e5dadc166ae51ff6f53fa60da7dca22afd65a31d44d8335f67c"],"state_sha256":"0995041cb4273f2beba0783c6a3289fe2b26b4e3fc9732320bcebc1253fec9fb"}