{"id":"W6887648048","doi":"10.17605/osf.io/498k6","title":"Assessing the research landscape and utility of LLMs in the clinical setting: protocol for a scoping review","year":2023,"lang":"en","type":"other","venue":"Open Science Framework","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Transformer; Natural language; Artificial neural network; Language model; Deep learning; Context (archaeology); Task (project management); Generative grammar","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","sts","scholarly_communication","open_science"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.1420901,0.0002551067,0.0008150508,0.0003065718,0.0006814382,0.001826259,0.007151004,0.0003418773,0.0003909757],"category_scores_gemma":[0.04901889,0.0001310694,0.0001031145,0.005704088,0.003653524,0.0004751203,0.002439449,0.001890431,0.0001201596],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004100432,"about_ca_system_score_gemma":0.002620713,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001794185,"about_ca_topic_score_gemma":0.0005355217,"domain_scores_codex":[0.9917454,0.00287001,0.001090821,0.001316092,0.002071258,0.0009064235],"domain_scores_gemma":[0.9843794,0.01151223,0.0009437758,0.002680426,0.0003603446,0.0001237855],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"systematic_review","study_design_scores_codex":[0.0000912376,0.0003565174,0.04044033,0.01608252,0.00002647527,0.00001046827,0.0009621923,6.720848e-7,0.000006966102,0.002329434,0.88034,0.05935323],"study_design_scores_gemma":[0.0008067584,0.0001972008,0.01633441,0.7422193,0.00003668435,0.000006986051,0.001479572,0.0005218513,0.000007518224,0.006165345,0.2317613,0.0004630941],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"protocol","genre_gemma":"protocol","genre_scores_codex":[0.00003225507,0.0011296,0.0001446167,0.003257908,0.0002007849,0.7801281,0.00004962564,0.0001008194,0.2149563],"genre_scores_gemma":[0.0005161647,0.0005246931,0.02563542,0.001105321,0.0006020049,0.9620351,0.000004600043,0.00101563,0.008561041],"genre_candidate":"protocol","genre_consensus":"protocol","teacher_disagreement_score":0.7261368,"threshold_uncertainty_score":0.9992099,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5550356977208956,"score_gpt":0.6782129478442712,"score_spread":0.1231772501233757,"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."}}