{"id":"W4401043314","doi":"10.18653/v1/2024.clinicalnlp-1.9","title":"Parameter-Efficient Fine-Tuning of LLaMA for the Clinical Domain","year":2024,"lang":"en","type":"article","venue":"","topic":"Medical Imaging Techniques and Applications","field":"Medicine","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"European Commission; Engineering and Physical Sciences Research Council; National Institute for Health and Care Research; Canadian Institute of Steel Construction; UK Research and Innovation; Nvidia; Accenture; Cisco Systems","keywords":"Computer science; Domain (mathematical analysis); Mathematics; Mathematical analysis","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006779664,0.00004911965,0.0001321844,0.00002431799,0.00003131697,0.00001074958,0.00007972342,0.00003996302,0.0001447762],"category_scores_gemma":[0.0002876739,0.00002500936,0.0001360049,0.0001153762,0.0001289443,0.000005417129,0.0000309293,0.0001326873,0.000009975364],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00000739353,"about_ca_system_score_gemma":0.00005024947,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008136004,"about_ca_topic_score_gemma":7.025187e-7,"domain_scores_codex":[0.9993649,0.00001224555,0.0002752334,0.0001372387,0.0001124134,0.0000979714],"domain_scores_gemma":[0.9983424,0.001280984,0.00002312608,0.0002561775,0.00003719645,0.00006013358],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00006431686,0.0005805795,0.0006882387,0.0003839675,0.0001697368,0.00001098748,0.0001836601,0.00001079053,0.007058101,0.2824565,0.4379699,0.2704231],"study_design_scores_gemma":[0.0004835786,0.000361049,0.0004847754,0.0002905716,0.0001790313,0.00002636322,0.0001201106,0.3168314,0.003586025,0.006715627,0.6708384,0.00008309274],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06923382,0.0003300342,0.874985,0.05104375,0.0001456941,0.000771053,0.00000727406,0.000209245,0.003274149],"genre_scores_gemma":[0.8356503,0.00003643301,0.1616522,0.0008388752,0.000185014,0.0001355458,0.000006918741,0.00001157752,0.001483078],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7664165,"threshold_uncertainty_score":0.1585198,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09146827919024167,"score_gpt":0.4403943549451226,"score_spread":0.3489260757548809,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}