{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HIKU56J37NPJS2XQXJWXMVF4IO","short_pith_number":"pith:HIKU56J3","schema_version":"1.0","canonical_sha256":"3a154ef93bfb5e996af0ba6d7654bc43a8006be3774ed3b2265eb5adbd13c87e","source":{"kind":"arxiv","id":"2403.08937","version":2},"attestation_state":"computed","paper":{"title":"Bugs in Large Language Models Generated Code: An Empirical Study","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Amin Nikanjam, Arghavan Moradi Dakhel, Florian Tambon, Foutse Khomh, Giuliano Antoniol, Michel C. Desmarais","submitted_at":"2024-03-13T20:12:01Z","abstract_excerpt":"Large Language Models (LLMs) for code have gained significant attention recently. They can generate code in different programming languages based on provided prompts, fulfilling a long-lasting dream in Software Engineering (SE), i.e., automatic code generation. Similar to human-written code, LLM-generated code is prone to bugs, and these bugs have not yet been thoroughly examined by the community. Given the increasing adoption of LLM-based code generation tools (e.g., GitHub Copilot) in SE activities, it is critical to understand the characteristics of bugs contained in code generated by LLMs."},"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":"2403.08937","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-03-13T20:12:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"61d0e062aea54a2a484628add665d33e7eb23c1529f5680bc99563a1b1969068","abstract_canon_sha256":"b2d12c66bf01b6fd3a71977d28994b6640f51abbd22545f6ef8b613733792419"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:57:30.448605Z","signature_b64":"5ebdQI6DF6Ymk64ZJ8scDMRO//VPYE4ZM/L3fFvxC98zmEiB+IYwI5QfiQGjvClFF35SQxBhN+D2AKc2ar0tAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a154ef93bfb5e996af0ba6d7654bc43a8006be3774ed3b2265eb5adbd13c87e","last_reissued_at":"2026-07-05T07:57:30.448082Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:57:30.448082Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bugs in Large Language Models Generated Code: An Empirical Study","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Amin Nikanjam, Arghavan Moradi Dakhel, Florian Tambon, Foutse Khomh, Giuliano Antoniol, Michel C. Desmarais","submitted_at":"2024-03-13T20:12:01Z","abstract_excerpt":"Large Language Models (LLMs) for code have gained significant attention recently. They can generate code in different programming languages based on provided prompts, fulfilling a long-lasting dream in Software Engineering (SE), i.e., automatic code generation. Similar to human-written code, LLM-generated code is prone to bugs, and these bugs have not yet been thoroughly examined by the community. Given the increasing adoption of LLM-based code generation tools (e.g., GitHub Copilot) in SE activities, it is critical to understand the characteristics of bugs contained in code generated by LLMs."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.08937","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/2403.08937/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":"2403.08937","created_at":"2026-07-05T07:57:30.448145+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.08937v2","created_at":"2026-07-05T07:57:30.448145+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.08937","created_at":"2026-07-05T07:57:30.448145+00:00"},{"alias_kind":"pith_short_12","alias_value":"HIKU56J37NPJ","created_at":"2026-07-05T07:57:30.448145+00:00"},{"alias_kind":"pith_short_16","alias_value":"HIKU56J37NPJS2XQ","created_at":"2026-07-05T07:57:30.448145+00:00"},{"alias_kind":"pith_short_8","alias_value":"HIKU56J3","created_at":"2026-07-05T07:57:30.448145+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.19102","citing_title":"Prompt Optimization for LLM Code Generation via Reinforcement Learning","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26672","citing_title":"What Makes Software Bugs Escape Testing? Evidence from a Large-Scale Empirical Study","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23361","citing_title":"An Empirical Evaluation of Locally Deployed LLMs for Bug Detection in Python Code","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05267","citing_title":"Bridging Generation and Training: A Systematic Review of Quality Issues in LLMs for Code","ref_index":116,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO","json":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO.json","graph_json":"https://pith.science/api/pith-number/HIKU56J37NPJS2XQXJWXMVF4IO/graph.json","events_json":"https://pith.science/api/pith-number/HIKU56J37NPJS2XQXJWXMVF4IO/events.json","paper":"https://pith.science/paper/HIKU56J3"},"agent_actions":{"view_html":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO","download_json":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO.json","view_paper":"https://pith.science/paper/HIKU56J3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.08937&json=true","fetch_graph":"https://pith.science/api/pith-number/HIKU56J37NPJS2XQXJWXMVF4IO/graph.json","fetch_events":"https://pith.science/api/pith-number/HIKU56J37NPJS2XQXJWXMVF4IO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO/action/storage_attestation","attest_author":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO/action/author_attestation","sign_citation":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO/action/citation_signature","submit_replication":"https://pith.science/pith/HIKU56J37NPJS2XQXJWXMVF4IO/action/replication_record"}},"created_at":"2026-07-05T07:57:30.448145+00:00","updated_at":"2026-07-05T07:57:30.448145+00:00"}