{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:W6ZPZWXE6A7R54UXQV6KHQ2UX4","short_pith_number":"pith:W6ZPZWXE","schema_version":"1.0","canonical_sha256":"b7b2fcdae4f03f1ef297857ca3c354bf350c6eb4fe53060e1bb609b68ee8ce1e","source":{"kind":"arxiv","id":"2305.10263","version":2},"attestation_state":"computed","paper":{"title":"M3KE: A Massive Multi-Level Multi-Subject Knowledge Evaluation Benchmark for Chinese Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chuang Liu, Deyi Xiong, Jianxiang Peng, Linhao Yu, Peiyi Zhang, Qingqing Lyu, Qun Liu, Renren Jin, Shuting Zhang, Tianyu Dong, Xiaohan Peng, Xiaowen Su, Yuqi Ren","submitted_at":"2023-05-17T14:56:31Z","abstract_excerpt":"Large language models have recently made tremendous progress in a variety of aspects, e.g., cross-task generalization, instruction following. Comprehensively evaluating the capability of large language models in multiple tasks is of great importance. In this paper, we propose M3KE, a Massive Multi-Level Multi-Subject Knowledge Evaluation benchmark, which is developed to measure knowledge acquired by Chinese large language models by testing their multitask accuracy in zero- and few-shot settings. We have collected 20,477 questions from 71 tasks. Our selection covers all major levels of Chinese "},"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":"2305.10263","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-17T14:56:31Z","cross_cats_sorted":[],"title_canon_sha256":"544a9012dcac68fb57ff47e27c9911300f422a37461d91eba027103558480988","abstract_canon_sha256":"c827e9e1b93ecae50c1098367b5ac4b5001a9702f9d8e265cb09c88954b28893"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:12:03.238144Z","signature_b64":"lwsfOAerZ23q7fqIwOgRX7s86Fd66reK16K3IIavNt2h/meiHu60VyVD2w8Bus6OM4hpcRPo/g3QeP8uWmX4Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b7b2fcdae4f03f1ef297857ca3c354bf350c6eb4fe53060e1bb609b68ee8ce1e","last_reissued_at":"2026-07-05T06:12:03.237757Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:12:03.237757Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"M3KE: A Massive Multi-Level Multi-Subject Knowledge Evaluation Benchmark for Chinese Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chuang Liu, Deyi Xiong, Jianxiang Peng, Linhao Yu, Peiyi Zhang, Qingqing Lyu, Qun Liu, Renren Jin, Shuting Zhang, Tianyu Dong, Xiaohan Peng, Xiaowen Su, Yuqi Ren","submitted_at":"2023-05-17T14:56:31Z","abstract_excerpt":"Large language models have recently made tremendous progress in a variety of aspects, e.g., cross-task generalization, instruction following. Comprehensively evaluating the capability of large language models in multiple tasks is of great importance. In this paper, we propose M3KE, a Massive Multi-Level Multi-Subject Knowledge Evaluation benchmark, which is developed to measure knowledge acquired by Chinese large language models by testing their multitask accuracy in zero- and few-shot settings. We have collected 20,477 questions from 71 tasks. Our selection covers all major levels of Chinese "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.10263","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/2305.10263/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":"2305.10263","created_at":"2026-07-05T06:12:03.237808+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.10263v2","created_at":"2026-07-05T06:12:03.237808+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.10263","created_at":"2026-07-05T06:12:03.237808+00:00"},{"alias_kind":"pith_short_12","alias_value":"W6ZPZWXE6A7R","created_at":"2026-07-05T06:12:03.237808+00:00"},{"alias_kind":"pith_short_16","alias_value":"W6ZPZWXE6A7R54UX","created_at":"2026-07-05T06:12:03.237808+00:00"},{"alias_kind":"pith_short_8","alias_value":"W6ZPZWXE","created_at":"2026-07-05T06:12:03.237808+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.09329","citing_title":"Towards Efficient and Effective Alignment of Large Language Models","ref_index":111,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4","json":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4.json","graph_json":"https://pith.science/api/pith-number/W6ZPZWXE6A7R54UXQV6KHQ2UX4/graph.json","events_json":"https://pith.science/api/pith-number/W6ZPZWXE6A7R54UXQV6KHQ2UX4/events.json","paper":"https://pith.science/paper/W6ZPZWXE"},"agent_actions":{"view_html":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4","download_json":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4.json","view_paper":"https://pith.science/paper/W6ZPZWXE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.10263&json=true","fetch_graph":"https://pith.science/api/pith-number/W6ZPZWXE6A7R54UXQV6KHQ2UX4/graph.json","fetch_events":"https://pith.science/api/pith-number/W6ZPZWXE6A7R54UXQV6KHQ2UX4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4/action/storage_attestation","attest_author":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4/action/author_attestation","sign_citation":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4/action/citation_signature","submit_replication":"https://pith.science/pith/W6ZPZWXE6A7R54UXQV6KHQ2UX4/action/replication_record"}},"created_at":"2026-07-05T06:12:03.237808+00:00","updated_at":"2026-07-05T06:12:03.237808+00:00"}