{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3G274RG4QB2FZOEGHOWH6FBOY5","short_pith_number":"pith:3G274RG4","schema_version":"1.0","canonical_sha256":"d9b5fe44dc80745cb8863bac7f142ec75b28b80bc40c4fb8e71da4c6a8001a77","source":{"kind":"arxiv","id":"2412.10056","version":1},"attestation_state":"computed","paper":{"title":"GAOKAO-Eval: Does high scores truly reflect strong capabilities in LLMs?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Changbo Wang, Chenchui Li, Hanglei Hu, Hang Yan, Jin Zhang, Linyang Li, Qipeng Guo, Tianyi Liang, Yunfan Shao, Yunhua Zhou, Zhikai Lei","submitted_at":"2024-12-13T11:38:10Z","abstract_excerpt":"Large Language Models (LLMs) are commonly evaluated using human-crafted benchmarks, under the premise that higher scores implicitly reflect stronger human-like performance. However, there is growing concern that LLMs may ``game\" these benchmarks due to data leakage, achieving high scores while struggling with tasks simple for humans. To substantively address the problem, we create GAOKAO-Eval, a comprehensive benchmark based on China's National College Entrance Examination (Gaokao), and conduct ``closed-book\" evaluations for representative models released prior to Gaokao. Contrary to prevailin"},"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":"2412.10056","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-13T11:38:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"437b5730c3f6f9af25d8a35ef00398f003241bcd0fd254321091f7a51bc984c7","abstract_canon_sha256":"9d81ed171b320c8675fd14d5a85e12430b3bc408b3dd76cda950858bc5b7991b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:48:46.374823Z","signature_b64":"QRjIrTgXry6qH3/1Ej8ssMFnzLiFEl1gMYqGRfOLTm8wDfuZhKND76OfF9QC90l8oJVO9X1Fo/h7TnsA3HGhDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9b5fe44dc80745cb8863bac7f142ec75b28b80bc40c4fb8e71da4c6a8001a77","last_reissued_at":"2026-07-05T09:48:46.374424Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:48:46.374424Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GAOKAO-Eval: Does high scores truly reflect strong capabilities in LLMs?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Changbo Wang, Chenchui Li, Hanglei Hu, Hang Yan, Jin Zhang, Linyang Li, Qipeng Guo, Tianyi Liang, Yunfan Shao, Yunhua Zhou, Zhikai Lei","submitted_at":"2024-12-13T11:38:10Z","abstract_excerpt":"Large Language Models (LLMs) are commonly evaluated using human-crafted benchmarks, under the premise that higher scores implicitly reflect stronger human-like performance. However, there is growing concern that LLMs may ``game\" these benchmarks due to data leakage, achieving high scores while struggling with tasks simple for humans. To substantively address the problem, we create GAOKAO-Eval, a comprehensive benchmark based on China's National College Entrance Examination (Gaokao), and conduct ``closed-book\" evaluations for representative models released prior to Gaokao. Contrary to prevailin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.10056","kind":"arxiv","version":1},"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/2412.10056/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":"2412.10056","created_at":"2026-07-05T09:48:46.374480+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.10056v1","created_at":"2026-07-05T09:48:46.374480+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.10056","created_at":"2026-07-05T09:48:46.374480+00:00"},{"alias_kind":"pith_short_12","alias_value":"3G274RG4QB2F","created_at":"2026-07-05T09:48:46.374480+00:00"},{"alias_kind":"pith_short_16","alias_value":"3G274RG4QB2FZOEG","created_at":"2026-07-05T09:48:46.374480+00:00"},{"alias_kind":"pith_short_8","alias_value":"3G274RG4","created_at":"2026-07-05T09:48:46.374480+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5","json":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5.json","graph_json":"https://pith.science/api/pith-number/3G274RG4QB2FZOEGHOWH6FBOY5/graph.json","events_json":"https://pith.science/api/pith-number/3G274RG4QB2FZOEGHOWH6FBOY5/events.json","paper":"https://pith.science/paper/3G274RG4"},"agent_actions":{"view_html":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5","download_json":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5.json","view_paper":"https://pith.science/paper/3G274RG4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.10056&json=true","fetch_graph":"https://pith.science/api/pith-number/3G274RG4QB2FZOEGHOWH6FBOY5/graph.json","fetch_events":"https://pith.science/api/pith-number/3G274RG4QB2FZOEGHOWH6FBOY5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5/action/storage_attestation","attest_author":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5/action/author_attestation","sign_citation":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5/action/citation_signature","submit_replication":"https://pith.science/pith/3G274RG4QB2FZOEGHOWH6FBOY5/action/replication_record"}},"created_at":"2026-07-05T09:48:46.374480+00:00","updated_at":"2026-07-05T09:48:46.374480+00:00"}