{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SRJWDHGQEMZHZQR4WT4RNRMGSQ","short_pith_number":"pith:SRJWDHGQ","schema_version":"1.0","canonical_sha256":"9453619cd023327cc23cb4f916c5869433b0fcc5ed63039c74344ae5892e48e0","source":{"kind":"arxiv","id":"2403.00894","version":2},"attestation_state":"computed","paper":{"title":"Comparing large language models and human programmers for generating programming code","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.PL"],"primary_cat":"cs.SE","authors_text":"Wenpin Hou, Zhicheng Ji","submitted_at":"2024-03-01T14:43:06Z","abstract_excerpt":"We systematically evaluated the performance of seven large language models in generating programming code using various prompt strategies, programming languages, and task difficulties. GPT-4 substantially outperforms other large language models, including Gemini Ultra and Claude 2. The coding performance of GPT-4 varies considerably with different prompt strategies. In most LeetCode and GeeksforGeeks coding contests evaluated in this study, GPT-4 employing the optimal prompt strategy outperforms 85 percent of human participants. Additionally, GPT-4 demonstrates strong capabilities in translati"},"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":"2403.00894","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2024-03-01T14:43:06Z","cross_cats_sorted":["cs.AI","cs.CL","cs.PL"],"title_canon_sha256":"a25d785926dddf87870b7d5fe1a64bdbc87bb228b734fd36ee6e87f9ee2df23c","abstract_canon_sha256":"bec5704648350d5cccecd1bd803c68449c80f412b8187bb9849fa555f9e0df35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:44.060437Z","signature_b64":"5j3zsO7RlG+mtisfsl63EK9piYGAZ9mBAej20iTFJ6I/RdiPNFXGRZZ8hG+wiB884i+ITyO4slrezVE4dOuJDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9453619cd023327cc23cb4f916c5869433b0fcc5ed63039c74344ae5892e48e0","last_reissued_at":"2026-07-05T10:02:44.059889Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:44.059889Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Comparing large language models and human programmers for generating programming code","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.PL"],"primary_cat":"cs.SE","authors_text":"Wenpin Hou, Zhicheng Ji","submitted_at":"2024-03-01T14:43:06Z","abstract_excerpt":"We systematically evaluated the performance of seven large language models in generating programming code using various prompt strategies, programming languages, and task difficulties. GPT-4 substantially outperforms other large language models, including Gemini Ultra and Claude 2. The coding performance of GPT-4 varies considerably with different prompt strategies. In most LeetCode and GeeksforGeeks coding contests evaluated in this study, GPT-4 employing the optimal prompt strategy outperforms 85 percent of human participants. Additionally, GPT-4 demonstrates strong capabilities in translati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.00894","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/2403.00894/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":"2403.00894","created_at":"2026-07-05T10:02:44.059946+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.00894v2","created_at":"2026-07-05T10:02:44.059946+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.00894","created_at":"2026-07-05T10:02:44.059946+00:00"},{"alias_kind":"pith_short_12","alias_value":"SRJWDHGQEMZH","created_at":"2026-07-05T10:02:44.059946+00:00"},{"alias_kind":"pith_short_16","alias_value":"SRJWDHGQEMZHZQR4","created_at":"2026-07-05T10:02:44.059946+00:00"},{"alias_kind":"pith_short_8","alias_value":"SRJWDHGQ","created_at":"2026-07-05T10:02:44.059946+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.08041","citing_title":"Quantifying the benefits of code hints for refactoring deprecated Java APIs","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ","json":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ.json","graph_json":"https://pith.science/api/pith-number/SRJWDHGQEMZHZQR4WT4RNRMGSQ/graph.json","events_json":"https://pith.science/api/pith-number/SRJWDHGQEMZHZQR4WT4RNRMGSQ/events.json","paper":"https://pith.science/paper/SRJWDHGQ"},"agent_actions":{"view_html":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ","download_json":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ.json","view_paper":"https://pith.science/paper/SRJWDHGQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.00894&json=true","fetch_graph":"https://pith.science/api/pith-number/SRJWDHGQEMZHZQR4WT4RNRMGSQ/graph.json","fetch_events":"https://pith.science/api/pith-number/SRJWDHGQEMZHZQR4WT4RNRMGSQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ/action/storage_attestation","attest_author":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ/action/author_attestation","sign_citation":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ/action/citation_signature","submit_replication":"https://pith.science/pith/SRJWDHGQEMZHZQR4WT4RNRMGSQ/action/replication_record"}},"created_at":"2026-07-05T10:02:44.059946+00:00","updated_at":"2026-07-05T10:02:44.059946+00:00"}