{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GPY3QQDI4ADGSHALB2IQJ2SPU2","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":"13a6e7db5c7be5f8e821318141ccc2824fbf93cef53e8e123112eb1b8d615a71","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-26T03:41:28Z","title_canon_sha256":"4ff76ff0c6fbfa55ea55455555c09dc0e507af24b7156fa0eb50dc645a2e3645"},"schema_version":"1.0","source":{"id":"2404.17140","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.17140","created_at":"2026-07-05T08:28:09Z"},{"alias_kind":"arxiv_version","alias_value":"2404.17140v2","created_at":"2026-07-05T08:28:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.17140","created_at":"2026-07-05T08:28:09Z"},{"alias_kind":"pith_short_12","alias_value":"GPY3QQDI4ADG","created_at":"2026-07-05T08:28:09Z"},{"alias_kind":"pith_short_16","alias_value":"GPY3QQDI4ADGSHAL","created_at":"2026-07-05T08:28:09Z"},{"alias_kind":"pith_short_8","alias_value":"GPY3QQDI","created_at":"2026-07-05T08:28:09Z"}],"graph_snapshots":[{"event_id":"sha256:155ffd59a206a6c60d81c373db6ad4c6660fb04f518dc75eb398bc36d1807a71","target":"graph","created_at":"2026-07-05T08:28:09Z","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/2404.17140/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Self-correction has emerged as a promising solution to boost the reasoning performance of large language models (LLMs), where LLMs refine their solutions using self-generated critiques that pinpoint the errors. This work explores whether small (<= 13B) language models (LMs) have the ability of self-correction on reasoning tasks with minimal inputs from stronger LMs. We propose a novel pipeline that prompts smaller LMs to collect self-correction data that supports the training of self-refinement abilities. First, we leverage correct solutions to guide the model in critiquing their incorrect res","authors_text":"Honglak Lee, Jaekyeom Kim, Lajanugen Logeswaran, Lu Wang, Moontae Lee, Muhammad Khalifa, Yunxiang Zhang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-26T03:41:28Z","title":"Small Language Models Need Strong Verifiers to Self-Correct Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.17140","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":"sha256:2003bd4f7ee2001c4c0ca1f27d57b8125d1fb7933f163847039df199e14a69ec","target":"record","created_at":"2026-07-05T08:28:09Z","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":"13a6e7db5c7be5f8e821318141ccc2824fbf93cef53e8e123112eb1b8d615a71","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-26T03:41:28Z","title_canon_sha256":"4ff76ff0c6fbfa55ea55455555c09dc0e507af24b7156fa0eb50dc645a2e3645"},"schema_version":"1.0","source":{"id":"2404.17140","kind":"arxiv","version":2}},"canonical_sha256":"33f1b84068e006691c0b0e9104ea4fa6bb4c4c6cb7523d57a0122a8c7775bbfc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"33f1b84068e006691c0b0e9104ea4fa6bb4c4c6cb7523d57a0122a8c7775bbfc","first_computed_at":"2026-07-05T08:28:09.526013Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:28:09.526013Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yKFU+c9BxyZ/NRCqCbjrmB+eGYJGhFy9esRXIy+c1+mhA1JlflFqPUa3fxEaPfAxUCNzzoiHmlfGCcC7HL2LDA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:28:09.526447Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.17140","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2003bd4f7ee2001c4c0ca1f27d57b8125d1fb7933f163847039df199e14a69ec","sha256:155ffd59a206a6c60d81c373db6ad4c6660fb04f518dc75eb398bc36d1807a71"],"state_sha256":"5efd36e71701896fd387c447c9529f1f8cc26b209f990e2b89a54931719309ca"}