{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:VFGX3NM5RI7H7X73EPPHKZHYDW","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"65be8b74e45b9784599b9dd7eec6ae5ebbaaa58bc136ad913e60747735140484","cross_cats_sorted":["cs.LG","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-07T17:52:51Z","title_canon_sha256":"0f88bfcef27c38e359b4d80d7d8a7b6d433034442e0359db59325c09e011d4c7"},"schema_version":"1.0","source":{"id":"2405.04520","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.04520","created_at":"2026-07-05T08:16:38Z"},{"alias_kind":"arxiv_version","alias_value":"2405.04520v1","created_at":"2026-07-05T08:16:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.04520","created_at":"2026-07-05T08:16:38Z"},{"alias_kind":"pith_short_12","alias_value":"VFGX3NM5RI7H","created_at":"2026-07-05T08:16:38Z"},{"alias_kind":"pith_short_16","alias_value":"VFGX3NM5RI7H7X73","created_at":"2026-07-05T08:16:38Z"},{"alias_kind":"pith_short_8","alias_value":"VFGX3NM5","created_at":"2026-07-05T08:16:38Z"}],"graph_snapshots":[{"event_id":"sha256:b6bd83d74f8726da51838c7fea0fa3f2558f84942cb7062d5a5a8a49261b37c7","target":"graph","created_at":"2026-07-05T08:16:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2405.04520/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have manifested strong ability to generate codes for productive activities. However, current benchmarks for code synthesis, such as HumanEval, MBPP, and DS-1000, are predominantly oriented towards introductory tasks on algorithm and data science, insufficiently satisfying challenging requirements prevalent in real-world coding. To fill this gap, we propose NaturalCodeBench (NCB), a challenging code benchmark designed to mirror the complexity and variety of scenarios in real coding tasks. NCB comprises 402 high-quality problems in Python and Java, meticulously selec","authors_text":"Hanlin Zhao, Jie Tang, Qinkai Zheng, Shudan Zhang, Xiaohan Zhang, Xiao Liu, Xiaotao Gu, Yuxiao Dong, Zehan Qi","cross_cats":["cs.LG","cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-07T17:52:51Z","title":"NaturalCodeBench: Examining Coding Performance Mismatch on HumanEval and Natural User Prompts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.04520","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ec6bb2ea9f70c6819f30ecebc0dc9afac637ebd14eea22a8a14b98261f1f58cd","target":"record","created_at":"2026-07-05T08:16:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"65be8b74e45b9784599b9dd7eec6ae5ebbaaa58bc136ad913e60747735140484","cross_cats_sorted":["cs.LG","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-07T17:52:51Z","title_canon_sha256":"0f88bfcef27c38e359b4d80d7d8a7b6d433034442e0359db59325c09e011d4c7"},"schema_version":"1.0","source":{"id":"2405.04520","kind":"arxiv","version":1}},"canonical_sha256":"a94d7db59d8a3e7fdffb23de7564f81db83cb95342fd579b21bb29732616f0a2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a94d7db59d8a3e7fdffb23de7564f81db83cb95342fd579b21bb29732616f0a2","first_computed_at":"2026-07-05T08:16:38.351520Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:16:38.351520Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vKOBMuNyxtEZbdcIwM/0ZRvbLycZZMHpGoRz8uXvVkYRA855VjmIaqt/QVz0ug88a1bN0d351CqhzDZlOUVmAw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:16:38.352083Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.04520","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ec6bb2ea9f70c6819f30ecebc0dc9afac637ebd14eea22a8a14b98261f1f58cd","sha256:b6bd83d74f8726da51838c7fea0fa3f2558f84942cb7062d5a5a8a49261b37c7"],"state_sha256":"3751ef6e053329e97f99882e7cef1ff7db07a774db8bc7556450551d15dd019c"}