{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KIQHG4W5BCUAAT2H7CUH4GYMXF","short_pith_number":"pith:KIQHG4W5","schema_version":"1.0","canonical_sha256":"52207372dd08a8004f47f8a87e1b0cb97806e73eac497ceee97c011e77380401","source":{"kind":"arxiv","id":"2402.13013","version":1},"attestation_state":"computed","paper":{"title":"Code Needs Comments: Enhancing Code LLMs with Comment Augmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dahua Lin, Demin Song, Hang Yan, Honglin Guo, Qipeng Guo, Shuhao Xing, Wenwei Zhang, Xipeng Qiu, Yudong Wang, Yunhua Zhou, Zifan Song","submitted_at":"2024-02-20T13:56:38Z","abstract_excerpt":"The programming skill is one crucial ability for Large Language Models (LLMs), necessitating a deep understanding of programming languages (PLs) and their correlation with natural languages (NLs). We examine the impact of pre-training data on code-focused LLMs' performance by assessing the comment density as a measure of PL-NL alignment. Given the scarcity of code-comment aligned data in pre-training corpora, we introduce a novel data augmentation method that generates comments for existing code, coupled with a data filtering strategy that filters out code data poorly correlated with natural l"},"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":"2402.13013","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-20T13:56:38Z","cross_cats_sorted":[],"title_canon_sha256":"1ceb33ea2fc331a2700276f9b8178f36ce3c295ae1d7dbbb59a93eef4f89b11f","abstract_canon_sha256":"d476ea7ecf6d57be103c1ed97b1c5d6b65d30ddb8c4552f769b5f5e3e84e3d9d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:47:19.655507Z","signature_b64":"fWkUNEY8625zJ1A0oHijRSPo2VKKJAtuMl5VCenmTTg3nKqjEUAOT8oZhk1tt6qQI8tVWDBfDB5SzdjtIBA3AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52207372dd08a8004f47f8a87e1b0cb97806e73eac497ceee97c011e77380401","last_reissued_at":"2026-07-05T07:47:19.654968Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:47:19.654968Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Code Needs Comments: Enhancing Code LLMs with Comment Augmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dahua Lin, Demin Song, Hang Yan, Honglin Guo, Qipeng Guo, Shuhao Xing, Wenwei Zhang, Xipeng Qiu, Yudong Wang, Yunhua Zhou, Zifan Song","submitted_at":"2024-02-20T13:56:38Z","abstract_excerpt":"The programming skill is one crucial ability for Large Language Models (LLMs), necessitating a deep understanding of programming languages (PLs) and their correlation with natural languages (NLs). We examine the impact of pre-training data on code-focused LLMs' performance by assessing the comment density as a measure of PL-NL alignment. Given the scarcity of code-comment aligned data in pre-training corpora, we introduce a novel data augmentation method that generates comments for existing code, coupled with a data filtering strategy that filters out code data poorly correlated with natural l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.13013","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/2402.13013/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":"2402.13013","created_at":"2026-07-05T07:47:19.655036+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.13013v1","created_at":"2026-07-05T07:47:19.655036+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.13013","created_at":"2026-07-05T07:47:19.655036+00:00"},{"alias_kind":"pith_short_12","alias_value":"KIQHG4W5BCUA","created_at":"2026-07-05T07:47:19.655036+00:00"},{"alias_kind":"pith_short_16","alias_value":"KIQHG4W5BCUAAT2H","created_at":"2026-07-05T07:47:19.655036+00:00"},{"alias_kind":"pith_short_8","alias_value":"KIQHG4W5","created_at":"2026-07-05T07:47:19.655036+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2403.17297","citing_title":"InternLM2 Technical Report","ref_index":110,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07957","citing_title":"Similar Pattern Annotation via Retrieval Knowledge for LLM-Based Test Code Fault Localization","ref_index":66,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF","json":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF.json","graph_json":"https://pith.science/api/pith-number/KIQHG4W5BCUAAT2H7CUH4GYMXF/graph.json","events_json":"https://pith.science/api/pith-number/KIQHG4W5BCUAAT2H7CUH4GYMXF/events.json","paper":"https://pith.science/paper/KIQHG4W5"},"agent_actions":{"view_html":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF","download_json":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF.json","view_paper":"https://pith.science/paper/KIQHG4W5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.13013&json=true","fetch_graph":"https://pith.science/api/pith-number/KIQHG4W5BCUAAT2H7CUH4GYMXF/graph.json","fetch_events":"https://pith.science/api/pith-number/KIQHG4W5BCUAAT2H7CUH4GYMXF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF/action/storage_attestation","attest_author":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF/action/author_attestation","sign_citation":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF/action/citation_signature","submit_replication":"https://pith.science/pith/KIQHG4W5BCUAAT2H7CUH4GYMXF/action/replication_record"}},"created_at":"2026-07-05T07:47:19.655036+00:00","updated_at":"2026-07-05T07:47:19.655036+00:00"}