{"id":"W4385572482","doi":"10.18653/v1/2023.bionlp-1.30","title":"Evaluation of ChatGPT on Biomedical Tasks: A Zero-Shot Comparison with Fine-Tuned Generative Transformers","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"Natural Sciences and Engineering Research Council of Canada; York University","keywords":"Automatic summarization; Computer science; Transformer; Generative grammar; Benchmark (surveying); Artificial intelligence; Relationship extraction; Domain (mathematical analysis); Language model; Training set; One shot; Machine learning; Natural language processing; Information extraction; Engineering; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001119706,0.0001110738,0.0001855801,0.0001896742,0.00005609794,0.0000276884,0.0002981132,0.00005196749,0.00005428872],"category_scores_gemma":[0.00003784799,0.0000792024,0.00003782401,0.0006604058,0.00005159116,0.0001542188,0.00003371648,0.00008953195,0.00003860457],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005292893,"about_ca_system_score_gemma":0.0001962737,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004959329,"about_ca_topic_score_gemma":0.00006348566,"domain_scores_codex":[0.9978302,0.0001162502,0.0002626388,0.0003214201,0.001262326,0.0002072052],"domain_scores_gemma":[0.9993322,0.00007105806,0.00006340901,0.000278982,0.0001804573,0.00007387648],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001096114,0.0004874382,0.0004736256,0.00005995019,0.0002406897,0.00001235718,0.0173533,0.05881173,0.02283001,0.06339775,0.005477113,0.8307464],"study_design_scores_gemma":[0.0008469659,0.0004404208,0.0004253581,0.00002976729,0.00002193175,0.000001374955,0.0001371252,0.9810958,0.0154584,0.001222406,0.0002115639,0.0001088968],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2349322,0.00001556898,0.759944,0.0018142,0.0001462237,0.000265104,0.000002148308,0.0001271237,0.002753469],"genre_scores_gemma":[0.9822893,0.000002707656,0.0174028,0.000111912,0.00003065302,0.00003439335,0.00001254052,0.000006077027,0.000109608],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9222841,"threshold_uncertainty_score":0.322978,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1115696166550883,"score_gpt":0.3477993786376039,"score_spread":0.2362297619825156,"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."}}