{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TNXECA7RPIPNC4WY4DUJMDONDX","short_pith_number":"pith:TNXECA7R","schema_version":"1.0","canonical_sha256":"9b6e4103f17a1ed172d8e0e8960dcd1dff08bfe4181beb837005ccdd31518b2b","source":{"kind":"arxiv","id":"2411.13611","version":3},"attestation_state":"computed","paper":{"title":"DSTC: Direct Preference Learning with Only Self-Generated Tests and Code to Improve Code LMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Boyi Liu, Shenao Zhang, Yingxiang Yang, Yongfei Liu, Zhaoran Wang, Zhihan Liu","submitted_at":"2024-11-20T02:03:16Z","abstract_excerpt":"Direct preference learning offers a promising and computation-efficient beyond supervised fine-tuning (SFT) for improving code generation in coding large language models (LMs). However, the scarcity of reliable preference data is a bottleneck for the performance of direct preference learning to improve the coding accuracy of code LMs. In this paper, we introduce \\underline{\\textbf{D}}irect Preference Learning with Only \\underline{\\textbf{S}}elf-Generated \\underline{\\textbf{T}}ests and \\underline{\\textbf{C}}ode (DSTC), a framework that leverages only self-generated code snippets and tests to co"},"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":"2411.13611","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-11-20T02:03:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2c0cfa04fed4894517b52ef5e92967963d66d96d0da27889d2de3260012b7bcc","abstract_canon_sha256":"dfbe655e4bbf98896726b8a4618a7f16a40b4eafd1494ba3d58d00f9b0702508"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:46:54.910589Z","signature_b64":"kAcHl/fOC4/fJp+imhNd4NbUqivmK/q5D4kR/G3d3w29Kq3Oq0l+2spqXniBe2BF4ileJTf8a7gHZ+mnJ3k/DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9b6e4103f17a1ed172d8e0e8960dcd1dff08bfe4181beb837005ccdd31518b2b","last_reissued_at":"2026-07-05T09:46:54.910017Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:46:54.910017Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DSTC: Direct Preference Learning with Only Self-Generated Tests and Code to Improve Code LMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Boyi Liu, Shenao Zhang, Yingxiang Yang, Yongfei Liu, Zhaoran Wang, Zhihan Liu","submitted_at":"2024-11-20T02:03:16Z","abstract_excerpt":"Direct preference learning offers a promising and computation-efficient beyond supervised fine-tuning (SFT) for improving code generation in coding large language models (LMs). However, the scarcity of reliable preference data is a bottleneck for the performance of direct preference learning to improve the coding accuracy of code LMs. In this paper, we introduce \\underline{\\textbf{D}}irect Preference Learning with Only \\underline{\\textbf{S}}elf-Generated \\underline{\\textbf{T}}ests and \\underline{\\textbf{C}}ode (DSTC), a framework that leverages only self-generated code snippets and tests to co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.13611","kind":"arxiv","version":3},"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/2411.13611/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":"2411.13611","created_at":"2026-07-05T09:46:54.910080+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.13611v3","created_at":"2026-07-05T09:46:54.910080+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.13611","created_at":"2026-07-05T09:46:54.910080+00:00"},{"alias_kind":"pith_short_12","alias_value":"TNXECA7RPIPN","created_at":"2026-07-05T09:46:54.910080+00:00"},{"alias_kind":"pith_short_16","alias_value":"TNXECA7RPIPNC4WY","created_at":"2026-07-05T09:46:54.910080+00:00"},{"alias_kind":"pith_short_8","alias_value":"TNXECA7R","created_at":"2026-07-05T09:46:54.910080+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16299","citing_title":"ACE: Self-Evolving LLM Coding Framework via Adversarial Unit Test Generation and Preference Optimization","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16299","citing_title":"ACE: Self-Evolving LLM Coding Framework via Adversarial Unit Test Generation and Preference Optimization","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07864","citing_title":"ZeroCoder: Can LLMs Improve Code Generation Without Ground-Truth Supervision?","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX","json":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX.json","graph_json":"https://pith.science/api/pith-number/TNXECA7RPIPNC4WY4DUJMDONDX/graph.json","events_json":"https://pith.science/api/pith-number/TNXECA7RPIPNC4WY4DUJMDONDX/events.json","paper":"https://pith.science/paper/TNXECA7R"},"agent_actions":{"view_html":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX","download_json":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX.json","view_paper":"https://pith.science/paper/TNXECA7R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.13611&json=true","fetch_graph":"https://pith.science/api/pith-number/TNXECA7RPIPNC4WY4DUJMDONDX/graph.json","fetch_events":"https://pith.science/api/pith-number/TNXECA7RPIPNC4WY4DUJMDONDX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX/action/storage_attestation","attest_author":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX/action/author_attestation","sign_citation":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX/action/citation_signature","submit_replication":"https://pith.science/pith/TNXECA7RPIPNC4WY4DUJMDONDX/action/replication_record"}},"created_at":"2026-07-05T09:46:54.910080+00:00","updated_at":"2026-07-05T09:46:54.910080+00:00"}