{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SRCGSO4TW5PGXHNCTH7KHTMUXN","short_pith_number":"pith:SRCGSO4T","schema_version":"1.0","canonical_sha256":"9444693b93b75e6b9da299fea3cd94bb7a4ac4c5453c06f62e125bfeeee982de","source":{"kind":"arxiv","id":"2503.10452","version":2},"attestation_state":"computed","paper":{"title":"DynaCode: A Dynamic Complexity-Aware Code Benchmark for Evaluating Large Language Models in Code Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chunchen Wei, Jinhao Duan, Kaidi Xu, Li Zhang, Wenhao Hu, Yue Zhang","submitted_at":"2025-03-13T15:18:56Z","abstract_excerpt":"The rapid advancement of large language models (LLMs) has significantly improved their performance in code generation tasks. However, existing code benchmarks remain static, consisting of fixed datasets with predefined problems. This makes them vulnerable to memorization during training, where LLMs recall specific test cases instead of generalizing to new problems, leading to data contamination and unreliable evaluation results. To address these issues, we introduce DynaCode, a dynamic, complexity-aware benchmark that overcomes the limitations of static datasets. DynaCode evaluates LLMs system"},"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":"2503.10452","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-13T15:18:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"16f3c7e1bb0f733c4303d8d9fd50b26134a15dce38b57832734626e70b71c9b4","abstract_canon_sha256":"f2b1759c119daefdce6536d0c23ac2242ddd93477333ecf42a7c5529a4a57129"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:42.981745Z","signature_b64":"v6vIvbZtjt5RjtInZJhQsuOXsEICiN6EH8n6zI0CSfVsR11zZH/sAHbnt3uC5gyCNL80FvD7n7ROUxCKvqaMCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9444693b93b75e6b9da299fea3cd94bb7a4ac4c5453c06f62e125bfeeee982de","last_reissued_at":"2026-07-05T11:11:42.981276Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:42.981276Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DynaCode: A Dynamic Complexity-Aware Code Benchmark for Evaluating Large Language Models in Code Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chunchen Wei, Jinhao Duan, Kaidi Xu, Li Zhang, Wenhao Hu, Yue Zhang","submitted_at":"2025-03-13T15:18:56Z","abstract_excerpt":"The rapid advancement of large language models (LLMs) has significantly improved their performance in code generation tasks. However, existing code benchmarks remain static, consisting of fixed datasets with predefined problems. This makes them vulnerable to memorization during training, where LLMs recall specific test cases instead of generalizing to new problems, leading to data contamination and unreliable evaluation results. To address these issues, we introduce DynaCode, a dynamic, complexity-aware benchmark that overcomes the limitations of static datasets. DynaCode evaluates LLMs system"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.10452","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/2503.10452/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":"2503.10452","created_at":"2026-07-05T11:11:42.981335+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.10452v2","created_at":"2026-07-05T11:11:42.981335+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.10452","created_at":"2026-07-05T11:11:42.981335+00:00"},{"alias_kind":"pith_short_12","alias_value":"SRCGSO4TW5PG","created_at":"2026-07-05T11:11:42.981335+00:00"},{"alias_kind":"pith_short_16","alias_value":"SRCGSO4TW5PGXHNC","created_at":"2026-07-05T11:11:42.981335+00:00"},{"alias_kind":"pith_short_8","alias_value":"SRCGSO4T","created_at":"2026-07-05T11:11:42.981335+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08009","citing_title":"From Execution to Education: A Bloom-Aligned Framework for Measuring Educational Control in LLMs","ref_index":74,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN","json":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN.json","graph_json":"https://pith.science/api/pith-number/SRCGSO4TW5PGXHNCTH7KHTMUXN/graph.json","events_json":"https://pith.science/api/pith-number/SRCGSO4TW5PGXHNCTH7KHTMUXN/events.json","paper":"https://pith.science/paper/SRCGSO4T"},"agent_actions":{"view_html":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN","download_json":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN.json","view_paper":"https://pith.science/paper/SRCGSO4T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.10452&json=true","fetch_graph":"https://pith.science/api/pith-number/SRCGSO4TW5PGXHNCTH7KHTMUXN/graph.json","fetch_events":"https://pith.science/api/pith-number/SRCGSO4TW5PGXHNCTH7KHTMUXN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN/action/storage_attestation","attest_author":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN/action/author_attestation","sign_citation":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN/action/citation_signature","submit_replication":"https://pith.science/pith/SRCGSO4TW5PGXHNCTH7KHTMUXN/action/replication_record"}},"created_at":"2026-07-05T11:11:42.981335+00:00","updated_at":"2026-07-05T11:11:42.981335+00:00"}