{"id":"W4378942418","doi":"10.48550/arxiv.2305.18486","title":"A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Automatic summarization; Benchmark (surveying); Computer science; Variety (cybernetics); Artificial intelligence; Strengths and weaknesses; Machine learning; Generative grammar; Data science; Machine translation; Benchmarking; Natural language processing; Psychology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01522117,0.004002182,0.002072486,0.005224337,0.002297186,0.00353779,0.005898026,0.003071869,0.005199947],"category_scores_gemma":[0.06210176,0.0009503012,0.002573367,0.005201782,0.001558605,0.008449485,0.005171049,0.00534093,0.005898688],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003480656,"about_ca_system_score_gemma":0.003521494,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01391201,"about_ca_topic_score_gemma":0.02613057,"domain_scores_codex":[0.9816234,0.010685,0.001283482,0.003360942,0.002522026,0.0005252479],"domain_scores_gemma":[0.9567909,0.02688871,0.001232968,0.008656266,0.00515684,0.001274351],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001907347,0.002139565,0.02617374,0.01024264,0.001741243,0.0007427966,0.002092537,0.08697131,0.01026054,0.008250643,0.3731408,0.4763368],"study_design_scores_gemma":[0.0008762745,0.002475886,0.02572346,0.001407457,0.0006935432,0.001715489,0.003170554,0.7370656,0.02416144,0.01657446,0.1856837,0.0004520767],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.3204849,0.02496678,0.2844597,0.007567813,0.003667291,0.006146987,0.1494324,0.1729045,0.03036962],"genre_scores_gemma":[0.3023639,0.003587525,0.2662845,0.002731507,0.0005956226,0.004682301,0.4060489,0.006240913,0.00746492],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01522117,"threshold_uncertainty_score":0.08049822,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2175232757801617,"score_gpt":0.2541113034357444,"score_spread":0.03658802765558267,"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."}}