{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:3UEC6X4JUDEXADXBQEKQBJH4JC","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":"0361354985e7eeb2d9c442e5c0fd04264162766294e59f3729327b5070fe3b77","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-30T17:59:07Z","title_canon_sha256":"6eb9811347cf7be805508823d1a8d54813ac7b5ecf590e3405120e788d54b364"},"schema_version":"1.0","source":{"id":"2407.00782","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.00782","created_at":"2026-07-05T08:43:56Z"},{"alias_kind":"arxiv_version","alias_value":"2407.00782v3","created_at":"2026-07-05T08:43:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00782","created_at":"2026-07-05T08:43:56Z"},{"alias_kind":"pith_short_12","alias_value":"3UEC6X4JUDEX","created_at":"2026-07-05T08:43:56Z"},{"alias_kind":"pith_short_16","alias_value":"3UEC6X4JUDEXADXB","created_at":"2026-07-05T08:43:56Z"},{"alias_kind":"pith_short_8","alias_value":"3UEC6X4J","created_at":"2026-07-05T08:43:56Z"}],"graph_snapshots":[{"event_id":"sha256:5172152158477414c85732fd5572dd87ccff184339c488f20ed4a2a8a5f72788","target":"graph","created_at":"2026-07-05T08:43:56Z","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/2407.00782/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Direct Preference Optimization (DPO) has proven effective at improving the performance of large language models (LLMs) on downstream tasks such as reasoning and alignment. In this work, we propose Step-Controlled DPO (SCDPO), a method for automatically providing stepwise error supervision by creating negative samples of mathematical reasoning rationales that start making errors at a specified step. By applying these samples in DPO training, SCDPO can better align the model to understand reasoning errors and output accurate reasoning steps. We apply SCDPO to both code-integrated and chain-of-th","authors_text":"Aojun Zhou, Hongsheng Li, Houxing Ren, Junting Pan, Ke Wang, Mingjie Zhan, Weikang Shi, Zimu Lu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-30T17:59:07Z","title":"Step-Controlled DPO: Leveraging Stepwise Error for Enhanced Mathematical Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00782","kind":"arxiv","version":3},"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":"sha256:7de77eb13bf0c04d20c36d6b28b7dac50c624794ab630475835fd403ef06543a","target":"record","created_at":"2026-07-05T08:43:56Z","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":"0361354985e7eeb2d9c442e5c0fd04264162766294e59f3729327b5070fe3b77","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-30T17:59:07Z","title_canon_sha256":"6eb9811347cf7be805508823d1a8d54813ac7b5ecf590e3405120e788d54b364"},"schema_version":"1.0","source":{"id":"2407.00782","kind":"arxiv","version":3}},"canonical_sha256":"dd082f5f89a0c9700ee1811500a4fc488d8cad1f12848a4336f40de14fa47bda","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"dd082f5f89a0c9700ee1811500a4fc488d8cad1f12848a4336f40de14fa47bda","first_computed_at":"2026-07-05T08:43:56.630003Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:43:56.630003Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"33bvjoe/L58yFy1NjqfpVNvOBrK0s0ZbKYhUpzVjP/ncuIen914Bxgpt3KumP+QUS588jIRQv1ePpo+6A+5mBA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:43:56.630512Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.00782","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7de77eb13bf0c04d20c36d6b28b7dac50c624794ab630475835fd403ef06543a","sha256:5172152158477414c85732fd5572dd87ccff184339c488f20ed4a2a8a5f72788"],"state_sha256":"8f482f4ee324fb713065559dfcd603a608526755fa4bae144f07ebe2eef34a59"}