{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:A55ZKWT5SNWCURMK6FGQUT5EDJ","short_pith_number":"pith:A55ZKWT5","schema_version":"1.0","canonical_sha256":"077b955a7d936c2a458af14d0a4fa41a6be131c6aee51e400024b09cd25e20e2","source":{"kind":"arxiv","id":"2307.12114","version":3},"attestation_state":"computed","paper":{"title":"A Zero-shot and Few-shot Study of Instruction-Finetuned Large Language Models Applied to Clinical and Biomedical Tasks","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Mickael Rouvier, Richard Dufour, Yanis Labrak","submitted_at":"2023-07-22T15:58:17Z","abstract_excerpt":"We evaluate four state-of-the-art instruction-tuned large language models (LLMs) -- ChatGPT, Flan-T5 UL2, Tk-Instruct, and Alpaca -- on a set of 13 real-world clinical and biomedical natural language processing (NLP) tasks in English, such as named-entity recognition (NER), question-answering (QA), relation extraction (RE), etc. Our overall results demonstrate that the evaluated LLMs begin to approach performance of state-of-the-art models in zero- and few-shot scenarios for most tasks, and particularly well for the QA task, even though they have never seen examples from these tasks before. Ho"},"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":"2307.12114","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2023-07-22T15:58:17Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"90abaeab4c8bbdff996dec30dedf1e523f054064522a6590b6a837186c4516c2","abstract_canon_sha256":"809dfc271b70806fcb63a0ace541e3f868f550a128d1d11b7e8f2aef16f8f4f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:16.580600Z","signature_b64":"fnlD8LTdLxKV2VzBrqm32O9Tzayiz1erc9VjaV8aHCbzr/FCO+B13spsKQCg+ORJ34Hc4l4FlJjD9R5AVWZTBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"077b955a7d936c2a458af14d0a4fa41a6be131c6aee51e400024b09cd25e20e2","last_reissued_at":"2026-07-05T08:29:16.580053Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:16.580053Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Zero-shot and Few-shot Study of Instruction-Finetuned Large Language Models Applied to Clinical and Biomedical Tasks","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Mickael Rouvier, Richard Dufour, Yanis Labrak","submitted_at":"2023-07-22T15:58:17Z","abstract_excerpt":"We evaluate four state-of-the-art instruction-tuned large language models (LLMs) -- ChatGPT, Flan-T5 UL2, Tk-Instruct, and Alpaca -- on a set of 13 real-world clinical and biomedical natural language processing (NLP) tasks in English, such as named-entity recognition (NER), question-answering (QA), relation extraction (RE), etc. Our overall results demonstrate that the evaluated LLMs begin to approach performance of state-of-the-art models in zero- and few-shot scenarios for most tasks, and particularly well for the QA task, even though they have never seen examples from these tasks before. Ho"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.12114","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/2307.12114/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":"2307.12114","created_at":"2026-07-05T08:29:16.580133+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.12114v3","created_at":"2026-07-05T08:29:16.580133+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.12114","created_at":"2026-07-05T08:29:16.580133+00:00"},{"alias_kind":"pith_short_12","alias_value":"A55ZKWT5SNWC","created_at":"2026-07-05T08:29:16.580133+00:00"},{"alias_kind":"pith_short_16","alias_value":"A55ZKWT5SNWCURMK","created_at":"2026-07-05T08:29:16.580133+00:00"},{"alias_kind":"pith_short_8","alias_value":"A55ZKWT5","created_at":"2026-07-05T08:29:16.580133+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.13177","citing_title":"Development of the user-friendly decision aid Rule-based Evaluation and Support Tool (REST) for optimizing the resources of an information extraction task","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ","json":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ.json","graph_json":"https://pith.science/api/pith-number/A55ZKWT5SNWCURMK6FGQUT5EDJ/graph.json","events_json":"https://pith.science/api/pith-number/A55ZKWT5SNWCURMK6FGQUT5EDJ/events.json","paper":"https://pith.science/paper/A55ZKWT5"},"agent_actions":{"view_html":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ","download_json":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ.json","view_paper":"https://pith.science/paper/A55ZKWT5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.12114&json=true","fetch_graph":"https://pith.science/api/pith-number/A55ZKWT5SNWCURMK6FGQUT5EDJ/graph.json","fetch_events":"https://pith.science/api/pith-number/A55ZKWT5SNWCURMK6FGQUT5EDJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ/action/storage_attestation","attest_author":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ/action/author_attestation","sign_citation":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ/action/citation_signature","submit_replication":"https://pith.science/pith/A55ZKWT5SNWCURMK6FGQUT5EDJ/action/replication_record"}},"created_at":"2026-07-05T08:29:16.580133+00:00","updated_at":"2026-07-05T08:29:16.580133+00:00"}