{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SE2QBASGPHLE2HLQSO2UFGMAGD","short_pith_number":"pith:SE2QBASG","schema_version":"1.0","canonical_sha256":"913500824679d64d1d7093b542998030c082ed3527d49dd05344d4251b7a56e1","source":{"kind":"arxiv","id":"2501.01264","version":2},"attestation_state":"computed","paper":{"title":"ProgCo: Program Helps Self-Correction of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bo Zheng, Jiaheng Liu, Weixun Wang, Wenbo Su, Xiaoshuai Song, Yanan Wu","submitted_at":"2025-01-02T13:59:20Z","abstract_excerpt":"Self-Correction aims to enable large language models (LLMs) to self-verify and self-refine their initial responses without external feedback. However, LLMs often fail to effectively self-verify and generate correct feedback, further misleading refinement and leading to the failure of self-correction, especially in complex reasoning tasks. In this paper, we propose Program-driven Self-Correction (ProgCo). First, program-driven verification (ProgVe) achieves complex verification logic and extensive validation through self-generated, self-executing verification pseudo-programs. Then, program-driv"},"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":"2501.01264","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-02T13:59:20Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a2d942ee559a237ae6aa035565821bfd2dacbdc20ef94169bf8c1ca2c3c26234","abstract_canon_sha256":"139241cbcc145a33f987a79d57c41df2ec411ad002c7b25f4da0643143b552b7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:03.730828Z","signature_b64":"VTwjrhCKB3XT1Qay27b8bP0iRAwxV7w8t3CKsE86D2Bo0Hyq1SBlnX/6NGa6jP4oeepA8qA1ncWkX5VVeC93Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"913500824679d64d1d7093b542998030c082ed3527d49dd05344d4251b7a56e1","last_reissued_at":"2026-07-05T11:10:03.730174Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:03.730174Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ProgCo: Program Helps Self-Correction of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bo Zheng, Jiaheng Liu, Weixun Wang, Wenbo Su, Xiaoshuai Song, Yanan Wu","submitted_at":"2025-01-02T13:59:20Z","abstract_excerpt":"Self-Correction aims to enable large language models (LLMs) to self-verify and self-refine their initial responses without external feedback. However, LLMs often fail to effectively self-verify and generate correct feedback, further misleading refinement and leading to the failure of self-correction, especially in complex reasoning tasks. In this paper, we propose Program-driven Self-Correction (ProgCo). First, program-driven verification (ProgVe) achieves complex verification logic and extensive validation through self-generated, self-executing verification pseudo-programs. Then, program-driv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.01264","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/2501.01264/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":"2501.01264","created_at":"2026-07-05T11:10:03.730248+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.01264v2","created_at":"2026-07-05T11:10:03.730248+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.01264","created_at":"2026-07-05T11:10:03.730248+00:00"},{"alias_kind":"pith_short_12","alias_value":"SE2QBASGPHLE","created_at":"2026-07-05T11:10:03.730248+00:00"},{"alias_kind":"pith_short_16","alias_value":"SE2QBASGPHLE2HLQ","created_at":"2026-07-05T11:10:03.730248+00:00"},{"alias_kind":"pith_short_8","alias_value":"SE2QBASG","created_at":"2026-07-05T11:10:03.730248+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.17419","citing_title":"From System 1 to System 2: A Survey of Reasoning Large Language Models","ref_index":221,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD","json":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD.json","graph_json":"https://pith.science/api/pith-number/SE2QBASGPHLE2HLQSO2UFGMAGD/graph.json","events_json":"https://pith.science/api/pith-number/SE2QBASGPHLE2HLQSO2UFGMAGD/events.json","paper":"https://pith.science/paper/SE2QBASG"},"agent_actions":{"view_html":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD","download_json":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD.json","view_paper":"https://pith.science/paper/SE2QBASG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.01264&json=true","fetch_graph":"https://pith.science/api/pith-number/SE2QBASGPHLE2HLQSO2UFGMAGD/graph.json","fetch_events":"https://pith.science/api/pith-number/SE2QBASGPHLE2HLQSO2UFGMAGD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD/action/storage_attestation","attest_author":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD/action/author_attestation","sign_citation":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD/action/citation_signature","submit_replication":"https://pith.science/pith/SE2QBASGPHLE2HLQSO2UFGMAGD/action/replication_record"}},"created_at":"2026-07-05T11:10:03.730248+00:00","updated_at":"2026-07-05T11:10:03.730248+00:00"}