{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:SW5TMW52DVPH3R7X7OM5N2DELQ","short_pith_number":"pith:SW5TMW52","canonical_record":{"source":{"id":"2306.03030","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-05T16:48:41Z","cross_cats_sorted":[],"title_canon_sha256":"f5522962a2101bb04e62ef141c3049caf614161302c9c44d006acae344bfdf86","abstract_canon_sha256":"3e8a5e3f1adf8fb913063df34e8467927d3ebca58ba7079bb4914d0ba2ba2b95"},"schema_version":"1.0"},"canonical_sha256":"95bb365bba1d5e7dc7f7fb99d6e8645c1071f23a5a48b27d2f2effaf8879458c","source":{"kind":"arxiv","id":"2306.03030","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.03030","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"arxiv_version","alias_value":"2306.03030v3","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.03030","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"pith_short_12","alias_value":"SW5TMW52DVPH","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"pith_short_16","alias_value":"SW5TMW52DVPH3R7X","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"pith_short_8","alias_value":"SW5TMW52","created_at":"2026-07-05T07:03:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:SW5TMW52DVPH3R7X7OM5N2DELQ","target":"record","payload":{"canonical_record":{"source":{"id":"2306.03030","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-05T16:48:41Z","cross_cats_sorted":[],"title_canon_sha256":"f5522962a2101bb04e62ef141c3049caf614161302c9c44d006acae344bfdf86","abstract_canon_sha256":"3e8a5e3f1adf8fb913063df34e8467927d3ebca58ba7079bb4914d0ba2ba2b95"},"schema_version":"1.0"},"canonical_sha256":"95bb365bba1d5e7dc7f7fb99d6e8645c1071f23a5a48b27d2f2effaf8879458c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:03:28.014377Z","signature_b64":"gXG/IcP32JX/HVqql3Sw66zme2nyuwz+EwqH+vk67vskBK4+CmglXJbAg+dZa6bRorB0ektbW8VakDcmVUsrAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"95bb365bba1d5e7dc7f7fb99d6e8645c1071f23a5a48b27d2f2effaf8879458c","last_reissued_at":"2026-07-05T07:03:28.013878Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:03:28.013878Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.03030","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:03:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"96x1Np7iHW1aD8rbja/kNgR9ZSyJBJhMV6CFzAFyBiJCbW5MWBxX3QWmqwm32You3HAXSoDLErsnda17VIkkAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T14:02:39.673426Z"},"content_sha256":"29e25f47a2d5734778f157e0c0dd8bb75d6ab2ff30a845b4bfad060c2ad267e4","schema_version":"1.0","event_id":"sha256:29e25f47a2d5734778f157e0c0dd8bb75d6ab2ff30a845b4bfad060c2ad267e4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:SW5TMW52DVPH3R7X7OM5N2DELQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking Large Language Models on CMExam -- A Comprehensive Chinese Medical Exam Dataset","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andrew Liu, Chenyu You, Dading Chong, Helin Wang, Junling Liu, Lei Zhu, Michael Lingzhi Li, Peilin Zhou, Yining Hua, Zhenhua Guo, Zhongyu Tian","submitted_at":"2023-06-05T16:48:41Z","abstract_excerpt":"Recent advancements in large language models (LLMs) have transformed the field of question answering (QA). However, evaluating LLMs in the medical field is challenging due to the lack of standardized and comprehensive datasets. To address this gap, we introduce CMExam, sourced from the Chinese National Medical Licensing Examination. CMExam consists of 60K+ multiple-choice questions for standardized and objective evaluations, as well as solution explanations for model reasoning evaluation in an open-ended manner. For in-depth analyses of LLMs, we invited medical professionals to label five addi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.03030","kind":"arxiv","version":3},"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/2306.03030/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:03:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZGKmgYjDe8PeSbGVFMo8wqev5Boi7xliSATCHzm2S46rj5nLjI/WpIdJlhPyGWnDKEYq/Nh3lUOXPjqxKdAQCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T14:02:39.673961Z"},"content_sha256":"a586aeecb8c4734ba71defacefd3bd6340a03c69c05992949d066e8a9ef9ea1a","schema_version":"1.0","event_id":"sha256:a586aeecb8c4734ba71defacefd3bd6340a03c69c05992949d066e8a9ef9ea1a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SW5TMW52DVPH3R7X7OM5N2DELQ/bundle.json","state_url":"https://pith.science/pith/SW5TMW52DVPH3R7X7OM5N2DELQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SW5TMW52DVPH3R7X7OM5N2DELQ/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":"2026-08-10T14:02:39Z","links":{"resolver":"https://pith.science/pith/SW5TMW52DVPH3R7X7OM5N2DELQ","bundle":"https://pith.science/pith/SW5TMW52DVPH3R7X7OM5N2DELQ/bundle.json","state":"https://pith.science/pith/SW5TMW52DVPH3R7X7OM5N2DELQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SW5TMW52DVPH3R7X7OM5N2DELQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:SW5TMW52DVPH3R7X7OM5N2DELQ","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":"3e8a5e3f1adf8fb913063df34e8467927d3ebca58ba7079bb4914d0ba2ba2b95","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-05T16:48:41Z","title_canon_sha256":"f5522962a2101bb04e62ef141c3049caf614161302c9c44d006acae344bfdf86"},"schema_version":"1.0","source":{"id":"2306.03030","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.03030","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"arxiv_version","alias_value":"2306.03030v3","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.03030","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"pith_short_12","alias_value":"SW5TMW52DVPH","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"pith_short_16","alias_value":"SW5TMW52DVPH3R7X","created_at":"2026-07-05T07:03:28Z"},{"alias_kind":"pith_short_8","alias_value":"SW5TMW52","created_at":"2026-07-05T07:03:28Z"}],"graph_snapshots":[{"event_id":"sha256:a586aeecb8c4734ba71defacefd3bd6340a03c69c05992949d066e8a9ef9ea1a","target":"graph","created_at":"2026-07-05T07:03:28Z","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/2306.03030/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in large language models (LLMs) have transformed the field of question answering (QA). However, evaluating LLMs in the medical field is challenging due to the lack of standardized and comprehensive datasets. To address this gap, we introduce CMExam, sourced from the Chinese National Medical Licensing Examination. CMExam consists of 60K+ multiple-choice questions for standardized and objective evaluations, as well as solution explanations for model reasoning evaluation in an open-ended manner. For in-depth analyses of LLMs, we invited medical professionals to label five addi","authors_text":"Andrew Liu, Chenyu You, Dading Chong, Helin Wang, Junling Liu, Lei Zhu, Michael Lingzhi Li, Peilin Zhou, Yining Hua, Zhenhua Guo, Zhongyu Tian","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-05T16:48:41Z","title":"Benchmarking Large Language Models on CMExam -- A Comprehensive Chinese Medical Exam Dataset"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.03030","kind":"arxiv","version":3},"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":"sha256:29e25f47a2d5734778f157e0c0dd8bb75d6ab2ff30a845b4bfad060c2ad267e4","target":"record","created_at":"2026-07-05T07:03:28Z","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":"3e8a5e3f1adf8fb913063df34e8467927d3ebca58ba7079bb4914d0ba2ba2b95","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-05T16:48:41Z","title_canon_sha256":"f5522962a2101bb04e62ef141c3049caf614161302c9c44d006acae344bfdf86"},"schema_version":"1.0","source":{"id":"2306.03030","kind":"arxiv","version":3}},"canonical_sha256":"95bb365bba1d5e7dc7f7fb99d6e8645c1071f23a5a48b27d2f2effaf8879458c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"95bb365bba1d5e7dc7f7fb99d6e8645c1071f23a5a48b27d2f2effaf8879458c","first_computed_at":"2026-07-05T07:03:28.013878Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:03:28.013878Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gXG/IcP32JX/HVqql3Sw66zme2nyuwz+EwqH+vk67vskBK4+CmglXJbAg+dZa6bRorB0ektbW8VakDcmVUsrAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:03:28.014377Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.03030","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:29e25f47a2d5734778f157e0c0dd8bb75d6ab2ff30a845b4bfad060c2ad267e4","sha256:a586aeecb8c4734ba71defacefd3bd6340a03c69c05992949d066e8a9ef9ea1a"],"state_sha256":"ab62af41457ab0177ff3249341b58aa8e9d4559765d2baead41b471c5df1e937"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uDCweQOJVNYLx/p86oG0MqtAj+ZtWn1Qsw0UkyuexjTdHi0o9k0D35LBJ6T+2UvjSCBHIkO+8tNWlbTqnosECg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T14:02:39.679073Z","bundle_sha256":"015902afc13e74997290f0652e7bc3a6d374f14138b071d56b3eceb6ffe1286f"}}