{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:AY7B3MSLIICKEHJXAFZ7D62B7J","short_pith_number":"pith:AY7B3MSL","canonical_record":{"source":{"id":"2306.09212","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-15T15:49:51Z","cross_cats_sorted":[],"title_canon_sha256":"85638d27497b05a8b6e6cfc2333eaaca9948e66c407771973cfd620a8c90441e","abstract_canon_sha256":"d9c30447ea56c65bd128e617eaa5ad9b4e951a27c9baad2ab43a562a40a37796"},"schema_version":"1.0"},"canonical_sha256":"063e1db24b4204a21d370173f1fb41fa4a41edf019b497bd0f153385a2b3ba48","source":{"kind":"arxiv","id":"2306.09212","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.09212","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"arxiv_version","alias_value":"2306.09212v2","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09212","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"pith_short_12","alias_value":"AY7B3MSLIICK","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"AY7B3MSLIICKEHJX","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"AY7B3MSL","created_at":"2026-05-18T12:33:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:AY7B3MSLIICKEHJXAFZ7D62B7J","target":"record","payload":{"canonical_record":{"source":{"id":"2306.09212","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-15T15:49:51Z","cross_cats_sorted":[],"title_canon_sha256":"85638d27497b05a8b6e6cfc2333eaaca9948e66c407771973cfd620a8c90441e","abstract_canon_sha256":"d9c30447ea56c65bd128e617eaa5ad9b4e951a27c9baad2ab43a562a40a37796"},"schema_version":"1.0"},"canonical_sha256":"063e1db24b4204a21d370173f1fb41fa4a41edf019b497bd0f153385a2b3ba48","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:46.588295Z","signature_b64":"k66DnZfs4bEnjYNQjeBZetig6fSkrRdVvjGp4NKFXffYv0dA5oS/kBtsQijvM9Lb60FaetuDx+juPyNngLB1DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"063e1db24b4204a21d370173f1fb41fa4a41edf019b497bd0f153385a2b3ba48","last_reissued_at":"2026-05-17T23:38:46.587873Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:46.587873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.09212","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PZqHZLM2aK2FGGUMb02tVV75ezhHMieBl4okyiJ9MJvlYVC3dKq2AIsMfTUhmeukmysM07zSEkC2uCQWn+2vCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-18T22:57:43.850784Z"},"content_sha256":"5aadda9c414455872ebcbc19c0e7560bb40567ad3373ded447219ca8a824ab9a","schema_version":"1.0","event_id":"sha256:5aadda9c414455872ebcbc19c0e7560bb40567ad3373ded447219ca8a824ab9a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:AY7B3MSLIICKEHJXAFZ7D62B7J","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CMMLU: Measuring massive multitask language understanding in Chinese","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"Most large language models score below 50 percent on a new Chinese multitask understanding benchmark.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fajri Koto, Hai Zhao, Haonan Li, Nan Duan, Timothy Baldwin, Yeyun Gong, Yifei Yang, Yixuan Zhang","submitted_at":"2023-06-15T15:49:51Z","abstract_excerpt":"As the capabilities of large language models (LLMs) continue to advance, evaluating their performance becomes increasingly crucial and challenging. This paper aims to bridge this gap by introducing CMMLU, a comprehensive Chinese benchmark that covers various subjects, including natural science, social sciences, engineering, and humanities. We conduct a thorough evaluation of 18 advanced multilingual- and Chinese-oriented LLMs, assessing their performance across different subjects and settings. The results reveal that most existing LLMs struggle to achieve an average accuracy of 50%, even when "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"most existing LLMs struggle to achieve an average accuracy of 50%, even when provided with in-context examples and chain-of-thought prompts, whereas the random baseline stands at 25%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The questions in CMMLU accurately represent the knowledge and reasoning demands of real Chinese-language tasks across the covered subjects.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CMMLU benchmark shows most advanced LLMs score below 50% accuracy on Chinese multitask understanding, well above the 25% random baseline but revealing major room for improvement.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Most large language models score below 50 percent on a new Chinese multitask understanding benchmark.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"afd9388db84a4af3ea61c6f5c90367cea2a2ecbba52536f98631f829755bf1fe"},"source":{"id":"2306.09212","kind":"arxiv","version":2},"verdict":{"id":"7e67a584-1eaf-49f9-b182-dc341b1f0120","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T20:58:16.649845Z","strongest_claim":"most existing LLMs struggle to achieve an average accuracy of 50%, even when provided with in-context examples and chain-of-thought prompts, whereas the random baseline stands at 25%.","one_line_summary":"CMMLU benchmark shows most advanced LLMs score below 50% accuracy on Chinese multitask understanding, well above the 25% random baseline but revealing major room for improvement.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The questions in CMMLU accurately represent the knowledge and reasoning demands of real Chinese-language tasks across the covered subjects.","pith_extraction_headline":"Most large language models score below 50 percent on a new Chinese multitask understanding benchmark."},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"9600d38a99db9286fc9a63dad72fb1478b226bdd6e157f3b3acdfdb5cd52bdde"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"7e67a584-1eaf-49f9-b182-dc341b1f0120"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"21XGJrdFtXr3K9PKrOXu1Vd6JO97N1xff08neSaFl2/05yGIydgqX/LRlpOwaebLCMptzv4LVsNclbOFZcVcAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-18T22:57:43.856862Z"},"content_sha256":"a4b3b9cc52e517db137e843ceddeed259f284b6a85329458d79ef4549a4d0970","schema_version":"1.0","event_id":"sha256:a4b3b9cc52e517db137e843ceddeed259f284b6a85329458d79ef4549a4d0970"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AY7B3MSLIICKEHJXAFZ7D62B7J/bundle.json","state_url":"https://pith.science/pith/AY7B3MSLIICKEHJXAFZ7D62B7J/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AY7B3MSLIICKEHJXAFZ7D62B7J/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-18T22:57:43Z","links":{"resolver":"https://pith.science/pith/AY7B3MSLIICKEHJXAFZ7D62B7J","bundle":"https://pith.science/pith/AY7B3MSLIICKEHJXAFZ7D62B7J/bundle.json","state":"https://pith.science/pith/AY7B3MSLIICKEHJXAFZ7D62B7J/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AY7B3MSLIICKEHJXAFZ7D62B7J/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:AY7B3MSLIICKEHJXAFZ7D62B7J","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":"d9c30447ea56c65bd128e617eaa5ad9b4e951a27c9baad2ab43a562a40a37796","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-15T15:49:51Z","title_canon_sha256":"85638d27497b05a8b6e6cfc2333eaaca9948e66c407771973cfd620a8c90441e"},"schema_version":"1.0","source":{"id":"2306.09212","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.09212","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"arxiv_version","alias_value":"2306.09212v2","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09212","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"pith_short_12","alias_value":"AY7B3MSLIICK","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"AY7B3MSLIICKEHJX","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"AY7B3MSL","created_at":"2026-05-18T12:33:33Z"}],"graph_snapshots":[{"event_id":"sha256:a4b3b9cc52e517db137e843ceddeed259f284b6a85329458d79ef4549a4d0970","target":"graph","created_at":"2026-05-17T23:38:46Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"most existing LLMs struggle to achieve an average accuracy of 50%, even when provided with in-context examples and chain-of-thought prompts, whereas the random baseline stands at 25%."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The questions in CMMLU accurately represent the knowledge and reasoning demands of real Chinese-language tasks across the covered subjects."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"CMMLU benchmark shows most advanced LLMs score below 50% accuracy on Chinese multitask understanding, well above the 25% random baseline but revealing major room for improvement."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Most large language models score below 50 percent on a new Chinese multitask understanding benchmark."}],"snapshot_sha256":"afd9388db84a4af3ea61c6f5c90367cea2a2ecbba52536f98631f829755bf1fe"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"9600d38a99db9286fc9a63dad72fb1478b226bdd6e157f3b3acdfdb5cd52bdde"},"paper":{"abstract_excerpt":"As the capabilities of large language models (LLMs) continue to advance, evaluating their performance becomes increasingly crucial and challenging. This paper aims to bridge this gap by introducing CMMLU, a comprehensive Chinese benchmark that covers various subjects, including natural science, social sciences, engineering, and humanities. We conduct a thorough evaluation of 18 advanced multilingual- and Chinese-oriented LLMs, assessing their performance across different subjects and settings. The results reveal that most existing LLMs struggle to achieve an average accuracy of 50%, even when ","authors_text":"Fajri Koto, Hai Zhao, Haonan Li, Nan Duan, Timothy Baldwin, Yeyun Gong, Yifei Yang, Yixuan Zhang","cross_cats":[],"headline":"Most large language models score below 50 percent on a new Chinese multitask understanding benchmark.","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-15T15:49:51Z","title":"CMMLU: Measuring massive multitask language understanding in Chinese"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.09212","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-16T20:58:16.649845Z","id":"7e67a584-1eaf-49f9-b182-dc341b1f0120","model_set":{"reader":"grok-4.3"},"one_line_summary":"CMMLU benchmark shows most advanced LLMs score below 50% accuracy on Chinese multitask understanding, well above the 25% random baseline but revealing major room for improvement.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Most large language models score below 50 percent on a new Chinese multitask understanding benchmark.","strongest_claim":"most existing LLMs struggle to achieve an average accuracy of 50%, even when provided with in-context examples and chain-of-thought prompts, whereas the random baseline stands at 25%.","weakest_assumption":"The questions in CMMLU accurately represent the knowledge and reasoning demands of real Chinese-language tasks across the covered subjects."}},"verdict_id":"7e67a584-1eaf-49f9-b182-dc341b1f0120"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5aadda9c414455872ebcbc19c0e7560bb40567ad3373ded447219ca8a824ab9a","target":"record","created_at":"2026-05-17T23:38:46Z","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":"d9c30447ea56c65bd128e617eaa5ad9b4e951a27c9baad2ab43a562a40a37796","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-15T15:49:51Z","title_canon_sha256":"85638d27497b05a8b6e6cfc2333eaaca9948e66c407771973cfd620a8c90441e"},"schema_version":"1.0","source":{"id":"2306.09212","kind":"arxiv","version":2}},"canonical_sha256":"063e1db24b4204a21d370173f1fb41fa4a41edf019b497bd0f153385a2b3ba48","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"063e1db24b4204a21d370173f1fb41fa4a41edf019b497bd0f153385a2b3ba48","first_computed_at":"2026-05-17T23:38:46.587873Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:46.587873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"k66DnZfs4bEnjYNQjeBZetig6fSkrRdVvjGp4NKFXffYv0dA5oS/kBtsQijvM9Lb60FaetuDx+juPyNngLB1DQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:46.588295Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.09212","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5aadda9c414455872ebcbc19c0e7560bb40567ad3373ded447219ca8a824ab9a","sha256:a4b3b9cc52e517db137e843ceddeed259f284b6a85329458d79ef4549a4d0970"],"state_sha256":"ee81be8248b48310e53b282667a07c9cd25fd9e79bc3b636c262fb49e597caf2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YONDTNS66e+k27h0eBbt9Jv5Pz/7/kvvLoJsKCLNoCIOK1anBV7JxGEEmAE8Zqs+hnkSNRk5eoBSEgYMO8NRAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-18T22:57:43.862263Z","bundle_sha256":"32d2ef31a209bef6a76a37b0d27a1c6184bcf31f5e18e2ea50d73c2eb6ce853b"}}