{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QTGUNLSYO3YVG75OLTSD7BHC4P","short_pith_number":"pith:QTGUNLSY","schema_version":"1.0","canonical_sha256":"84cd46ae5876f1537fae5ce43f84e2e3edef4722efe1659b6ebd016fcd123f63","source":{"kind":"arxiv","id":"2406.15673","version":2},"attestation_state":"computed","paper":{"title":"Large Language Models have Intrinsic Self-Correction Ability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amir Nassereldine, ChangJae Lee, Chenhui Xu, Dancheng Liu, Jiajie Li, Jinjun Xiong, Ruiyang Qin, Utkarsh Kumar, Yiyu Shi, Yuting Hu, Ziming Yang","submitted_at":"2024-06-21T22:29:40Z","abstract_excerpt":"Large language models (LLMs) have attracted significant attention for their exceptional abilities in various natural language processing tasks, but they suffer from hallucinations that will cause performance degradation. One promising solution to improve the LLMs' performance is to ask LLMs to revise their answer after generation, a technique known as self-correction. Among the two types of self-correction, intrinsic self-correction is considered a promising direction because it does not utilize external knowledge. However, recent works doubt the validity of LLM's ability to conduct intrinsic "},"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":"2406.15673","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-21T22:29:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"af4d5fbc7d69ba28030a4f680a9403b00a3002a1b9aa3b34745fc48d79261ba7","abstract_canon_sha256":"4875570eecad0fc80cdb8e7d103a0e4ac8f90d40cb918cde0e8633e4b7f72512"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:53:02.417636Z","signature_b64":"hCL6ZbBlohNa3F0UiFqHZ2sVXH70UKCL7OYRWq98kXVpmMQeCDbRwk7w6ieGbew5jK4D7R0xwaUfevuyk4UhBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"84cd46ae5876f1537fae5ce43f84e2e3edef4722efe1659b6ebd016fcd123f63","last_reissued_at":"2026-07-05T09:53:02.417200Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:53:02.417200Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models have Intrinsic Self-Correction Ability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amir Nassereldine, ChangJae Lee, Chenhui Xu, Dancheng Liu, Jiajie Li, Jinjun Xiong, Ruiyang Qin, Utkarsh Kumar, Yiyu Shi, Yuting Hu, Ziming Yang","submitted_at":"2024-06-21T22:29:40Z","abstract_excerpt":"Large language models (LLMs) have attracted significant attention for their exceptional abilities in various natural language processing tasks, but they suffer from hallucinations that will cause performance degradation. One promising solution to improve the LLMs' performance is to ask LLMs to revise their answer after generation, a technique known as self-correction. Among the two types of self-correction, intrinsic self-correction is considered a promising direction because it does not utilize external knowledge. However, recent works doubt the validity of LLM's ability to conduct intrinsic "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.15673","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/2406.15673/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":"2406.15673","created_at":"2026-07-05T09:53:02.417254+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.15673v2","created_at":"2026-07-05T09:53:02.417254+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.15673","created_at":"2026-07-05T09:53:02.417254+00:00"},{"alias_kind":"pith_short_12","alias_value":"QTGUNLSYO3YV","created_at":"2026-07-05T09:53:02.417254+00:00"},{"alias_kind":"pith_short_16","alias_value":"QTGUNLSYO3YVG75O","created_at":"2026-07-05T09:53:02.417254+00:00"},{"alias_kind":"pith_short_8","alias_value":"QTGUNLSY","created_at":"2026-07-05T09:53:02.417254+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01023","citing_title":"Evidence-Supported Credit Risk Report Generation Using News-Centric Financial Knowledge Graphs","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27564","citing_title":"The Future of Facts: Tracing the Factual Generation-Verification Gap","ref_index":128,"is_internal_anchor":false},{"citing_arxiv_id":"2510.00072","citing_title":"Unlocking Zero-Shot Geospatial Reasoning via Indirect Rewards","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2601.00514","citing_title":"The Illusion of Insight in Reasoning Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22271","citing_title":"How LLMs Detect and Correct Their Own Errors: The Role of Internal Confidence Signals","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P","json":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P.json","graph_json":"https://pith.science/api/pith-number/QTGUNLSYO3YVG75OLTSD7BHC4P/graph.json","events_json":"https://pith.science/api/pith-number/QTGUNLSYO3YVG75OLTSD7BHC4P/events.json","paper":"https://pith.science/paper/QTGUNLSY"},"agent_actions":{"view_html":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P","download_json":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P.json","view_paper":"https://pith.science/paper/QTGUNLSY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.15673&json=true","fetch_graph":"https://pith.science/api/pith-number/QTGUNLSYO3YVG75OLTSD7BHC4P/graph.json","fetch_events":"https://pith.science/api/pith-number/QTGUNLSYO3YVG75OLTSD7BHC4P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P/action/storage_attestation","attest_author":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P/action/author_attestation","sign_citation":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P/action/citation_signature","submit_replication":"https://pith.science/pith/QTGUNLSYO3YVG75OLTSD7BHC4P/action/replication_record"}},"created_at":"2026-07-05T09:53:02.417254+00:00","updated_at":"2026-07-05T09:53:02.417254+00:00"}