{"id":"W6910493271","doi":"10.48448/hn1y-fx85","title":"Have You Seen That Number? Investigating Extrapolation in Question Answering Models","year":2021,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Extrapolation; Question answering; Key (lock); Comprehension; Reading (process); Numerical models","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":[0.004011354,0.001015755,0.000655931,0.001146484,0.0003760465,0.002210637,0.001425399,0.001401349,0.006153692],"category_scores_gemma":[0.03014719,0.0003856717,0.001203786,0.0007959757,0.00105627,0.00507151,0.001498389,0.002747455,0.001734343],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001090458,"about_ca_system_score_gemma":0.0007002892,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006020565,"about_ca_topic_score_gemma":0.005369792,"domain_scores_codex":[0.9980172,0.001212177,0.00008299821,0.0004734063,0.0001443262,0.00006981169],"domain_scores_gemma":[0.9785796,0.01870051,0.0006849889,0.0009694471,0.0007625564,0.0003027403],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001148252,0.0004706429,0.04974817,0.001555178,0.0004538049,0.0008695725,0.005717878,0.364651,0.006245233,0.1368853,0.02456213,0.4076929],"study_design_scores_gemma":[0.00003418178,0.00008926263,0.003132506,0.0001619,0.00004851117,0.0001296116,0.0002716419,0.8777273,0.001191635,0.1121083,0.005074569,0.00003062037],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"other","genre_scores_codex":[0.3870402,0.007950673,0.5511811,0.01151867,0.0003674818,0.0002088089,0.004569664,0.005404666,0.03175874],"genre_scores_gemma":[0.9429898,0.001037764,0.04723852,0.0009132981,0.0001555997,0.00011442,0.003711854,0.0001962168,0.0036425],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.006153692,"threshold_uncertainty_score":0.02121431,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07817537908674883,"score_gpt":0.3423284876348174,"score_spread":0.2641531085480686,"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."}}