{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001704657,0.0001047963,0.00009004534,0.00009532482,0.0002226522,0.000774877,0.0002056665,0.000032653,0.00002960349],"category_scores_gemma":[0.0006401949,0.00009690991,0.00002161835,0.0003286944,0.00001569259,0.001271681,0.00008151884,0.00009931235,0.0001067519],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001414958,"about_ca_system_score_gemma":0.0002199734,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003025889,"about_ca_topic_score_gemma":0.000005727326,"domain_scores_codex":[0.9985257,0.0002062378,0.0001985759,0.0004062213,0.0005169924,0.0001462632],"domain_scores_gemma":[0.9987654,0.0004283536,0.00004776786,0.0002811834,0.0003633639,0.0001139023],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00007012863,0.0004982131,0.0003934867,0.0002686988,0.0001461469,0.000009297942,0.01374296,0.001700214,0.1939696,0.6359912,0.03557215,0.1176379],"study_design_scores_gemma":[0.00007562586,0.00006630782,0.003512979,0.0002037378,0.00003635377,0.0001331102,0.0002806189,0.02679518,0.00274914,0.005684999,0.9601875,0.0002744761],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04994125,0.02207053,0.5330088,0.3808956,0.005057779,0.001035916,0.000006668714,0.0002831639,0.007700377],"genre_scores_gemma":[0.8433148,0.0001380662,0.154156,0.001317128,0.000539774,0.00002584427,0.000006179279,0.00001019621,0.0004920173],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9246153,"threshold_uncertainty_score":0.7472157,"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."}}