{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4E7XVPTLXRT6UY3BT2ZE4RMFU4","short_pith_number":"pith:4E7XVPTL","schema_version":"1.0","canonical_sha256":"e13f7abe6bbc67ea63619eb24e4585a700c716acae27ce1404f19c3539002d5f","source":{"kind":"arxiv","id":"2412.19513","version":1},"attestation_state":"computed","paper":{"title":"Confidence v.s. Critique: A Decomposition of Self-Correction Capability for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Junyang Lin, Yichang Zhang, Yudong Wang, Zhe Yang, Zhifang Sui, Ziyao Xu","submitted_at":"2024-12-27T08:09:11Z","abstract_excerpt":"Large Language Models (LLMs) can correct their self-generated responses, but a decline in accuracy after self-correction is also witnessed. To have a deeper understanding of self-correction, we endeavor to decompose, evaluate, and analyze the self-correction behaviors of LLMs. By enumerating and analyzing answer correctness before and after self-correction, we decompose the self-correction capability into confidence (being confident to correct answers) and critique (turning wrong answers to correct) capabilities, and propose two metrics from a probabilistic perspective to measure these 2 capab"},"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":"2412.19513","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-27T08:09:11Z","cross_cats_sorted":[],"title_canon_sha256":"66c14bb212a80dd1fb3417963a7fd4275c9b4ea01481d5ff0a89fb1d4f548751","abstract_canon_sha256":"82410e7be0a37f6c3e9dfe03afad4415efb506b92250272da14d59727933673b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:54:37.066949Z","signature_b64":"aMiK3ou9JcY/0+IzIqFtwZMb8kpLT2fyCbutrJ642Ff9DOLAd0a52pi5f9+S3gFM1zbr8P4dFUsNTfIARkGOBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e13f7abe6bbc67ea63619eb24e4585a700c716acae27ce1404f19c3539002d5f","last_reissued_at":"2026-07-05T09:54:37.066452Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:54:37.066452Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Confidence v.s. Critique: A Decomposition of Self-Correction Capability for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Junyang Lin, Yichang Zhang, Yudong Wang, Zhe Yang, Zhifang Sui, Ziyao Xu","submitted_at":"2024-12-27T08:09:11Z","abstract_excerpt":"Large Language Models (LLMs) can correct their self-generated responses, but a decline in accuracy after self-correction is also witnessed. To have a deeper understanding of self-correction, we endeavor to decompose, evaluate, and analyze the self-correction behaviors of LLMs. By enumerating and analyzing answer correctness before and after self-correction, we decompose the self-correction capability into confidence (being confident to correct answers) and critique (turning wrong answers to correct) capabilities, and propose two metrics from a probabilistic perspective to measure these 2 capab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.19513","kind":"arxiv","version":1},"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/2412.19513/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":"2412.19513","created_at":"2026-07-05T09:54:37.066515+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.19513v1","created_at":"2026-07-05T09:54:37.066515+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.19513","created_at":"2026-07-05T09:54:37.066515+00:00"},{"alias_kind":"pith_short_12","alias_value":"4E7XVPTLXRT6","created_at":"2026-07-05T09:54:37.066515+00:00"},{"alias_kind":"pith_short_16","alias_value":"4E7XVPTLXRT6UY3B","created_at":"2026-07-05T09:54:37.066515+00:00"},{"alias_kind":"pith_short_8","alias_value":"4E7XVPTL","created_at":"2026-07-05T09:54:37.066515+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.06805","citing_title":"Cognitive Loop of Thought: Reversible Hierarchical Markov Chain for Efficient Mathematical Reasoning","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4","json":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4.json","graph_json":"https://pith.science/api/pith-number/4E7XVPTLXRT6UY3BT2ZE4RMFU4/graph.json","events_json":"https://pith.science/api/pith-number/4E7XVPTLXRT6UY3BT2ZE4RMFU4/events.json","paper":"https://pith.science/paper/4E7XVPTL"},"agent_actions":{"view_html":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4","download_json":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4.json","view_paper":"https://pith.science/paper/4E7XVPTL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.19513&json=true","fetch_graph":"https://pith.science/api/pith-number/4E7XVPTLXRT6UY3BT2ZE4RMFU4/graph.json","fetch_events":"https://pith.science/api/pith-number/4E7XVPTLXRT6UY3BT2ZE4RMFU4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4/action/storage_attestation","attest_author":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4/action/author_attestation","sign_citation":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4/action/citation_signature","submit_replication":"https://pith.science/pith/4E7XVPTLXRT6UY3BT2ZE4RMFU4/action/replication_record"}},"created_at":"2026-07-05T09:54:37.066515+00:00","updated_at":"2026-07-05T09:54:37.066515+00:00"}