{"id":"W2963241005","doi":"10.1145/3322640.3326711","title":"A Reliable and Accurate Multiple Choice Question Answering System for Due Diligence","year":2019,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Cisco Systems (Canada)","funders":"","keywords":"Question answering; Computer science; Due diligence; Scarcity; Artificial intelligence; Classifier (UML); Task (project management); Machine learning; Oversampling; Bandwidth (computing); Finance","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":[0.003984063,0.0009424471,0.001597618,0.002103213,0.001023476,0.001730328,0.002099779,0.003163832,0.00578265],"category_scores_gemma":[0.01359957,0.0003506095,0.0007253264,0.001284634,0.0004078677,0.003758744,0.001708233,0.00178014,0.005941562],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001003763,"about_ca_system_score_gemma":0.001464132,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003613643,"about_ca_topic_score_gemma":0.003108896,"domain_scores_codex":[0.9968353,0.0007772231,0.0002889807,0.0009741228,0.0009096365,0.0002146777],"domain_scores_gemma":[0.9925026,0.003409779,0.0005461411,0.0008961972,0.002268547,0.0003766615],"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":[0.001001043,0.001272097,0.01480561,0.0006328588,0.0001610256,0.0008491563,0.001165775,0.01856472,0.07764192,0.01037087,0.1021916,0.7713433],"study_design_scores_gemma":[0.0001140618,0.0002764598,0.006946339,0.00005449426,0.00007222078,0.0005364918,0.0003216529,0.9096209,0.03759063,0.01238847,0.03196229,0.0001158248],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.093364,0.0009359475,0.8123381,0.002525192,0.0004384577,0.001047468,0.005134393,0.07796198,0.006254574],"genre_scores_gemma":[0.4505839,0.0002849099,0.5254179,0.0008567972,0.0004183792,0.0008233745,0.01408578,0.0005499203,0.006979011],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00578265,"threshold_uncertainty_score":0.02107,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02117962379379888,"score_gpt":0.2575918210204401,"score_spread":0.2364121972266412,"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."}}