{"id":"W4402081416","doi":"10.1007/978-3-031-70563-2_10","title":"Analyzing Biases in Popular Answer Selection Datasets on Neural-Based QA Models","year":2024,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Computer science; Selection (genetic algorithm); Artificial intelligence; Machine learning; Artificial neural network; Natural language processing","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01165453,0.0006938587,0.0008511336,0.002140737,0.0007148873,0.001570715,0.001321909,0.001631523,0.002770685],"category_scores_gemma":[0.0589851,0.0002945339,0.0008565378,0.00249227,0.0006724803,0.002750251,0.00128288,0.001874855,0.001263673],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001525786,"about_ca_system_score_gemma":0.0007543629,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005672724,"about_ca_topic_score_gemma":0.00857961,"domain_scores_codex":[0.9947488,0.003201865,0.0003356417,0.0006830801,0.0007980915,0.0002325679],"domain_scores_gemma":[0.930404,0.05882964,0.001660648,0.004300749,0.004187864,0.0006170205],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.01119837,0.002361983,0.2389833,0.002788839,0.001996928,0.0005033033,0.002317967,0.1886428,0.01447738,0.01758007,0.1433728,0.3757763],"study_design_scores_gemma":[0.0003466256,0.0005794082,0.04992962,0.0002269792,0.0003221881,0.0002834171,0.0006718336,0.9118737,0.007305996,0.01935513,0.009024906,0.00008021158],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.960194,0.00384262,0.01831139,0.001504693,0.0002363788,0.0001008901,0.01118468,0.001333796,0.003291665],"genre_scores_gemma":[0.9638591,0.0003834483,0.009891319,0.0002492788,0.0001531371,0.00008322005,0.0235007,0.0001689627,0.001710745],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9883454,"threshold_uncertainty_score":0.06163573,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04471806634163544,"score_gpt":0.2792266015176221,"score_spread":0.2345085351759866,"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."}}