{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:4IYB7I4PPBQTU55F3MIY43XWRJ","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":"576ad311f8e9660588bf3109ffab31a8959547e2ea0d824208e8e0a02529fc1e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-28T03:03:51Z","title_canon_sha256":"d336cd26d3595e5bbbf32923a3d00a20f14f5dfd8d40a55529dfc04489556d20"},"schema_version":"1.0","source":{"id":"2306.15887","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.15887","created_at":"2026-07-05T06:25:45Z"},{"alias_kind":"arxiv_version","alias_value":"2306.15887v1","created_at":"2026-07-05T06:25:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.15887","created_at":"2026-07-05T06:25:45Z"},{"alias_kind":"pith_short_12","alias_value":"4IYB7I4PPBQT","created_at":"2026-07-05T06:25:45Z"},{"alias_kind":"pith_short_16","alias_value":"4IYB7I4PPBQTU55F","created_at":"2026-07-05T06:25:45Z"},{"alias_kind":"pith_short_8","alias_value":"4IYB7I4P","created_at":"2026-07-05T06:25:45Z"}],"graph_snapshots":[{"event_id":"sha256:2a9fbb99b8d0e150afd0a679f406096934ea2be4c270ffa07804a4cbd593323f","target":"graph","created_at":"2026-07-05T06:25:45Z","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/2306.15887/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The use of large language models (LLMs) in healthcare is gaining popularity, but their practicality and safety in clinical settings have not been thoroughly assessed. In high-stakes environments like medical settings, trust and safety are critical issues for LLMs. To address these concerns, we present an approach to evaluate the performance and trustworthiness of a GPT3.5 model for medical image protocol assignment. We compare it with a fine-tuned BERT model and a radiologist. In addition, we have a radiologist review the GPT3.5 output to evaluate its decision-making process. Our evaluation da","authors_text":"Elizabeth Tong, Mohammad R. K. Mofrad, Salmonn Talebi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-28T03:03:51Z","title":"Beyond the Hype: Assessing the Performance, Trustworthiness, and Clinical Suitability of GPT3.5"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.15887","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:ffddebc38b408f188b2896b5a5a4b1b06bb79906110747107ba3a0b2484a2250","target":"record","created_at":"2026-07-05T06:25:45Z","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":"576ad311f8e9660588bf3109ffab31a8959547e2ea0d824208e8e0a02529fc1e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-28T03:03:51Z","title_canon_sha256":"d336cd26d3595e5bbbf32923a3d00a20f14f5dfd8d40a55529dfc04489556d20"},"schema_version":"1.0","source":{"id":"2306.15887","kind":"arxiv","version":1}},"canonical_sha256":"e2301fa38f78613a77a5db118e6ef68a6db9ce9ad7150054a3715069bbd2b6ff","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e2301fa38f78613a77a5db118e6ef68a6db9ce9ad7150054a3715069bbd2b6ff","first_computed_at":"2026-07-05T06:25:45.930518Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:25:45.930518Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hrZYkVHH2ED9ofvFJYsq79hwTMdhVYzck1hcLgoxA7qwVoYWtKui+ADfgVRAin5/N7HaVFVqV9tJBk8KX5kKCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:25:45.930990Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.15887","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ffddebc38b408f188b2896b5a5a4b1b06bb79906110747107ba3a0b2484a2250","sha256:2a9fbb99b8d0e150afd0a679f406096934ea2be4c270ffa07804a4cbd593323f"],"state_sha256":"56ff2e32b515069013d029f657dd8bf7d9eba19bd75cedad261ba3906b195fc1"}