{"id":"W4416673024","doi":"10.1007/s10462-025-11421-5","title":"Exploring unanswerability in machine reading comprehension: approaches, benchmarks, and open challenges","year":2025,"lang":"en","type":"article","venue":"Artificial Intelligence Review","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University; Ted Rogers Centre for Heart Research; University of Guelph","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Key (lock); Comprehension; Reading (process); Work (physics); Reading comprehension","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":[0.001830859,0.0001785893,0.0004791855,0.0001095981,0.0001114549,0.0001558468,0.001187222,0.00004253128,0.00001971272],"category_scores_gemma":[0.0002211255,0.0001627589,0.00004944491,0.0005926876,0.00005347136,0.0008556214,0.001090279,0.0002353615,0.00001096129],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006401924,"about_ca_system_score_gemma":0.00006444008,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002590732,"about_ca_topic_score_gemma":0.0002668568,"domain_scores_codex":[0.997902,0.000249007,0.000683463,0.0007314636,0.0001515835,0.000282462],"domain_scores_gemma":[0.9987341,0.0002420798,0.00008460493,0.0008226696,0.00004493059,0.00007165697],"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":[0.000001632376,0.00003322287,0.00007846443,0.0005854408,0.000003702129,0.000003311221,0.0002245518,0.00004594678,0.000007203641,0.3272519,0.00000474911,0.6717599],"study_design_scores_gemma":[0.0001295805,0.0001292917,0.002169494,0.02660563,0.00006300833,0.00003584692,0.0009017567,0.5147109,0.002565732,0.4019082,0.04950982,0.001270737],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"review","genre_gemma":"empirical","genre_scores_codex":[0.004229912,0.535751,0.4188947,0.02708926,0.0006900133,0.002081939,0.000001369636,0.0001270862,0.01113469],"genre_scores_gemma":[0.5579834,0.4103097,0.03066378,0.0007468415,0.00003594098,0.0002228302,0.000003669655,0.000008975128,0.00002489173],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.6704891,"threshold_uncertainty_score":0.6637115,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5294704198234306,"score_gpt":0.3733258316109949,"score_spread":0.1561445882124357,"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."}}