{"id":"W4327498768","doi":"10.1007/978-3-031-28241-6_11","title":"Pre-processing Matters! Improved Wikipedia Corpora for Open-Domain Question Answering","year":2023,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Question answering; Natural language processing; Information retrieval; Domain (mathematical analysis); Landmark; Artificial intelligence; Text corpus; Open domain; Pipeline (software)","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":[0.002610521,0.001449152,0.00123885,0.005171527,0.001711614,0.002869511,0.001675995,0.001277636,0.02822721],"category_scores_gemma":[0.01761835,0.001052899,0.001060559,0.004221028,0.0005688489,0.005161494,0.003187682,0.002865486,0.02843975],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006451901,"about_ca_system_score_gemma":0.002021072,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006529718,"about_ca_topic_score_gemma":0.01480823,"domain_scores_codex":[0.9973012,0.0008495923,0.0003073984,0.0007399878,0.0006245174,0.0001773001],"domain_scores_gemma":[0.9830422,0.006757754,0.0004644684,0.002964402,0.006096799,0.000674371],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004008083,0.0004847806,0.002793842,0.00222792,0.0001826773,0.0004007947,0.001055291,0.002958035,0.02892032,0.006701498,0.6379682,0.3159058],"study_design_scores_gemma":[0.0002665482,0.0002363187,0.01216881,0.0005128859,0.0003181034,0.0007226663,0.001605982,0.05759106,0.05819633,0.01777111,0.8503751,0.0002352113],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.089596,0.007430704,0.310394,0.004558771,0.005320698,0.001732398,0.4018745,0.1118242,0.06726865],"genre_scores_gemma":[0.08103187,0.001245496,0.2485175,0.0005350276,0.0007862792,0.001243249,0.6391909,0.007877875,0.01957181],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.02822721,"threshold_uncertainty_score":0.09442943,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02880047406840642,"score_gpt":0.2781804076201266,"score_spread":0.2493799335517202,"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."}}