{"id":"W4391227797","doi":"10.1111/emip.12590","title":"Using OpenAI GPT to Generate Reading Comprehension Items","year":2024,"lang":"en","type":"article","venue":"Educational Measurement Issues and Practice","topic":"Text Readability and Simplification","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reading comprehension; Comprehension; Reading (process); Sentence; Computer science; Cognition; Natural language processing; Artificial intelligence; Sample (material); Psychology; Cognitive psychology; Linguistics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006708799,0.001260643,0.000686163,0.002250435,0.0003782295,0.001606593,0.001627094,0.0008019766,0.008932563],"category_scores_gemma":[0.05775053,0.0005497162,0.0008059987,0.001459246,0.0005323419,0.001530407,0.001752133,0.001273373,0.004859292],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006136245,"about_ca_system_score_gemma":0.0006781245,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008953472,"about_ca_topic_score_gemma":0.0008328997,"domain_scores_codex":[0.9941532,0.002851601,0.0008039345,0.0008391111,0.001242512,0.0001096561],"domain_scores_gemma":[0.9454595,0.03644957,0.001921354,0.006327393,0.009467513,0.0003746814],"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":[0.0007058989,0.0009847138,0.01967767,0.0007102253,0.0001121132,0.0006234278,0.004586733,0.01089695,0.02845694,0.003003902,0.01057994,0.9196615],"study_design_scores_gemma":[0.001631092,0.006623516,0.127087,0.0008522757,0.0004307322,0.004053501,0.0035609,0.4151379,0.3243577,0.02106772,0.09449054,0.0007070538],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2399371,0.0001321561,0.70674,0.0001827952,0.0002588954,0.008523776,0.004437896,0.02691454,0.01287306],"genre_scores_gemma":[0.2596103,0.00009997491,0.7200114,0.00007469645,0.00004361029,0.008080788,0.006155544,0.002069866,0.003853862],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008932563,"threshold_uncertainty_score":0.03547996,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2292006550062576,"score_gpt":0.4155313631435855,"score_spread":0.1863307081373279,"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."}}