{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GGPIPIH4YHOUOF5QUXUAPUDT43","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":"0b4416d89f5a31feeb88d1a99df40aa9807d9dc7756b2ded1187fcb1e3d44031","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-05T04:15:07Z","title_canon_sha256":"892df28e4234b5483a88b956e228e100c10c5c1079dfaed7c821ddd0d8a16dcf"},"schema_version":"1.0","source":{"id":"2406.02919","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.02919","created_at":"2026-07-05T08:27:32Z"},{"alias_kind":"arxiv_version","alias_value":"2406.02919v1","created_at":"2026-07-05T08:27:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.02919","created_at":"2026-07-05T08:27:32Z"},{"alias_kind":"pith_short_12","alias_value":"GGPIPIH4YHOU","created_at":"2026-07-05T08:27:32Z"},{"alias_kind":"pith_short_16","alias_value":"GGPIPIH4YHOUOF5Q","created_at":"2026-07-05T08:27:32Z"},{"alias_kind":"pith_short_8","alias_value":"GGPIPIH4","created_at":"2026-07-05T08:27:32Z"}],"graph_snapshots":[{"event_id":"sha256:f80c0f794714e7a7fd6bdaa3cdba339228ff141304030e41c0d7d8b2c4031de9","target":"graph","created_at":"2026-07-05T08:27:32Z","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/2406.02919/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have excelled across domains, also delivering notable performance on the medical evaluation benchmarks, such as MedQA. However, there still exists a significant gap between the reported performance and the practical effectiveness in real-world medical scenarios. In this paper, we aim to explore the causes of this gap by employing a multifaceted examination schema to systematically probe the actual mastery of medical knowledge by current LLMs. Specifically, we develop a novel evaluation framework MultifacetEval to examine the degree and coverage of LLMs in encoding ","authors_text":"Chen Ning, Ji Wu, Xien Liu, Yuxuan Zhou","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-05T04:15:07Z","title":"MultifacetEval: Multifaceted Evaluation to Probe LLMs in Mastering Medical Knowledge"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.02919","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":"sha256:8dca1d9979406fc0f8efb2fd2d7edc7ef7dfaf16a835f42736ea9a6a75931663","target":"record","created_at":"2026-07-05T08:27:32Z","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":"0b4416d89f5a31feeb88d1a99df40aa9807d9dc7756b2ded1187fcb1e3d44031","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-05T04:15:07Z","title_canon_sha256":"892df28e4234b5483a88b956e228e100c10c5c1079dfaed7c821ddd0d8a16dcf"},"schema_version":"1.0","source":{"id":"2406.02919","kind":"arxiv","version":1}},"canonical_sha256":"319e87a0fcc1dd4717b0a5e807d073e6d6e92d903c2202456ae65106a9ed233e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"319e87a0fcc1dd4717b0a5e807d073e6d6e92d903c2202456ae65106a9ed233e","first_computed_at":"2026-07-05T08:27:32.687132Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:27:32.687132Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"o/AllqPJ1T5zY7FOg+xIlpbkQOFihXLODgmW5VnS1++Z5exQymwCH+U9dE/22crDGC+49n7QyFIUQHBv/nlDCw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:27:32.687609Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.02919","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8dca1d9979406fc0f8efb2fd2d7edc7ef7dfaf16a835f42736ea9a6a75931663","sha256:f80c0f794714e7a7fd6bdaa3cdba339228ff141304030e41c0d7d8b2c4031de9"],"state_sha256":"9883ee26457aa4dd1e609544b74215ee7534f25ac065e7e0404ea7988d78417c"}