{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TSBLBHVICZHATTAP6SVF73OD55","short_pith_number":"pith:TSBLBHVI","schema_version":"1.0","canonical_sha256":"9c82b09ea8164e09cc0ff4aa5fedc3ef5eaf4a4c57c18a406a416020a5cba303","source":{"kind":"arxiv","id":"2403.18746","version":1},"attestation_state":"computed","paper":{"title":"CYCLE: Learning to Self-Refine the Code Generation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Baishakhi Ray, Gail Kaiser, Marcus J. Min, Yangruibo Ding","submitted_at":"2024-03-27T16:45:02Z","abstract_excerpt":"Pre-trained code language models have achieved promising performance in code generation and improved the programming efficiency of human developers. However, their self-refinement capability is typically overlooked by the existing evaluations of code LMs, which focus only on the accuracy of the one-time prediction. For the cases when code LMs fail to implement the correct program, developers actually find it hard to debug and fix the faulty prediction since it is not written by the developers themselves. Unfortunately, our study reveals that code LMs cannot efficiently self-refine their faulty"},"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":"2403.18746","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SE","submitted_at":"2024-03-27T16:45:02Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ece9319c58229c3c781668e1bbcf3ace51601e1625079ccbe6bfab70d5d620cf","abstract_canon_sha256":"bb88735bdda9090ee8c68ccdcfd2a5a893c3c703f3c218485bebd7c1d35517e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:01:26.458340Z","signature_b64":"B4B5PvhykvDHuuxxmZ0OuMQk2+MfoYSjslzIRqwbdiWmZ79wt5XsZGYnHaSHnjhyb28wlZ+DaBzJpFORHlrLBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c82b09ea8164e09cc0ff4aa5fedc3ef5eaf4a4c57c18a406a416020a5cba303","last_reissued_at":"2026-07-05T08:01:26.457880Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:01:26.457880Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CYCLE: Learning to Self-Refine the Code Generation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Baishakhi Ray, Gail Kaiser, Marcus J. Min, Yangruibo Ding","submitted_at":"2024-03-27T16:45:02Z","abstract_excerpt":"Pre-trained code language models have achieved promising performance in code generation and improved the programming efficiency of human developers. However, their self-refinement capability is typically overlooked by the existing evaluations of code LMs, which focus only on the accuracy of the one-time prediction. For the cases when code LMs fail to implement the correct program, developers actually find it hard to debug and fix the faulty prediction since it is not written by the developers themselves. Unfortunately, our study reveals that code LMs cannot efficiently self-refine their faulty"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.18746","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/2403.18746/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":"2403.18746","created_at":"2026-07-05T08:01:26.457932+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.18746v1","created_at":"2026-07-05T08:01:26.457932+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.18746","created_at":"2026-07-05T08:01:26.457932+00:00"},{"alias_kind":"pith_short_12","alias_value":"TSBLBHVICZHA","created_at":"2026-07-05T08:01:26.457932+00:00"},{"alias_kind":"pith_short_16","alias_value":"TSBLBHVICZHATTAP","created_at":"2026-07-05T08:01:26.457932+00:00"},{"alias_kind":"pith_short_8","alias_value":"TSBLBHVI","created_at":"2026-07-05T08:01:26.457932+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31511","citing_title":"Falsification, Not Exposure: An Internally Preregistered Placebo-Controlled Decomposition of Self-Repair Feedback in Frozen Small Code Models","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55","json":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55.json","graph_json":"https://pith.science/api/pith-number/TSBLBHVICZHATTAP6SVF73OD55/graph.json","events_json":"https://pith.science/api/pith-number/TSBLBHVICZHATTAP6SVF73OD55/events.json","paper":"https://pith.science/paper/TSBLBHVI"},"agent_actions":{"view_html":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55","download_json":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55.json","view_paper":"https://pith.science/paper/TSBLBHVI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.18746&json=true","fetch_graph":"https://pith.science/api/pith-number/TSBLBHVICZHATTAP6SVF73OD55/graph.json","fetch_events":"https://pith.science/api/pith-number/TSBLBHVICZHATTAP6SVF73OD55/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55/action/storage_attestation","attest_author":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55/action/author_attestation","sign_citation":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55/action/citation_signature","submit_replication":"https://pith.science/pith/TSBLBHVICZHATTAP6SVF73OD55/action/replication_record"}},"created_at":"2026-07-05T08:01:26.457932+00:00","updated_at":"2026-07-05T08:01:26.457932+00:00"}