{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GAQB47LGMEUWTRVCU2H5HVNOQ6","short_pith_number":"pith:GAQB47LG","schema_version":"1.0","canonical_sha256":"30201e7d66612969c6a2a68fd3d5ae878e62443450266aa6ccd39b8c194f75b7","source":{"kind":"arxiv","id":"2406.08731","version":3},"attestation_state":"computed","paper":{"title":"Towards Understanding the Characteristics of Code Generation Errors Made by Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Da Song, Lei Ma, Shengmai Chen, Tianyi Zhang, Yuheng Huang, Zhijie Wang, Zijie Zhou","submitted_at":"2024-06-13T01:29:52Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated unprecedented capabilities in code generation. However, there remains a limited understanding of code generation errors that LLMs can produce. To bridge the gap, we conducted an in-depth analysis of code generation errors across six representative LLMs on the HumanEval dataset. Specifically, we first employed open coding and thematic analysis to distill a comprehensive taxonomy of code generation errors. We analyzed two dimensions of error characteristics -- semantic characteristics and syntactic characteristics. Our analysis revealed that LLMs of"},"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":"2406.08731","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-06-13T01:29:52Z","cross_cats_sorted":[],"title_canon_sha256":"b8fe662eb2b74bdc21916e4c544248101599a296f83de1a1329300114f71301b","abstract_canon_sha256":"10aa667ba08332087176fce4e02e927886d144cec79ae7dc1dc0cf812916580b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:26.506814Z","signature_b64":"rORl/XKArsEdNN6cMpv/ng79V3YIYgFg5SH5VKeA1u5/zWKjWRshoYweJEa/w3qb3ayUEyKjhsIURkQbD8amBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"30201e7d66612969c6a2a68fd3d5ae878e62443450266aa6ccd39b8c194f75b7","last_reissued_at":"2026-07-05T10:13:26.506271Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:26.506271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Understanding the Characteristics of Code Generation Errors Made by Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Da Song, Lei Ma, Shengmai Chen, Tianyi Zhang, Yuheng Huang, Zhijie Wang, Zijie Zhou","submitted_at":"2024-06-13T01:29:52Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated unprecedented capabilities in code generation. However, there remains a limited understanding of code generation errors that LLMs can produce. To bridge the gap, we conducted an in-depth analysis of code generation errors across six representative LLMs on the HumanEval dataset. Specifically, we first employed open coding and thematic analysis to distill a comprehensive taxonomy of code generation errors. We analyzed two dimensions of error characteristics -- semantic characteristics and syntactic characteristics. Our analysis revealed that LLMs of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.08731","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/2406.08731/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":"2406.08731","created_at":"2026-07-05T10:13:26.506337+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.08731v3","created_at":"2026-07-05T10:13:26.506337+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.08731","created_at":"2026-07-05T10:13:26.506337+00:00"},{"alias_kind":"pith_short_12","alias_value":"GAQB47LGMEUW","created_at":"2026-07-05T10:13:26.506337+00:00"},{"alias_kind":"pith_short_16","alias_value":"GAQB47LGMEUWTRVC","created_at":"2026-07-05T10:13:26.506337+00:00"},{"alias_kind":"pith_short_8","alias_value":"GAQB47LG","created_at":"2026-07-05T10:13:26.506337+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.17029","citing_title":"Task Abstention for Large Language Models in Code Generation","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2512.19644","citing_title":"A survey of generative AI adoption and perceived productivity among scientists who program","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03245","citing_title":"FVRuleLearner: Operator-Level Reasoning Tree (Op-Tree)-Based Rules Learning for Formal Verification","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10598","citing_title":"Budget-Efficient Automatic Algorithm Design via Code Graph","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02431","citing_title":"ARIADNE: Agentic Reward-Informed Adaptive Decision Exploration via Blackboard-Driven MCTS for Competitive Program Generation","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6","json":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6.json","graph_json":"https://pith.science/api/pith-number/GAQB47LGMEUWTRVCU2H5HVNOQ6/graph.json","events_json":"https://pith.science/api/pith-number/GAQB47LGMEUWTRVCU2H5HVNOQ6/events.json","paper":"https://pith.science/paper/GAQB47LG"},"agent_actions":{"view_html":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6","download_json":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6.json","view_paper":"https://pith.science/paper/GAQB47LG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.08731&json=true","fetch_graph":"https://pith.science/api/pith-number/GAQB47LGMEUWTRVCU2H5HVNOQ6/graph.json","fetch_events":"https://pith.science/api/pith-number/GAQB47LGMEUWTRVCU2H5HVNOQ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6/action/storage_attestation","attest_author":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6/action/author_attestation","sign_citation":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6/action/citation_signature","submit_replication":"https://pith.science/pith/GAQB47LGMEUWTRVCU2H5HVNOQ6/action/replication_record"}},"created_at":"2026-07-05T10:13:26.506337+00:00","updated_at":"2026-07-05T10:13:26.506337+00:00"}