{"id":"W2799424953","doi":"10.1162/tacl_a_00018","title":"Questionable Answers in Question Answering Research: Reproducibility and Variability of Published Results","year":2018,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Expert finding and Q&A systems","field":"Computer Science","cited_by":73,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Reproducibility; Computer science; Popularity; Reliability (semiconductor); Sample (material); Field (mathematics); Range (aeronautics); Code (set theory); Artificial intelligence; Data science; Statistics; Psychology; Mathematics; Social psychology","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":["metaresearch"],"category_scores_codex":[0.5940362,0.001063335,0.002159154,0.009942214,0.00283769,0.01190427,0.006442023,0.004593777,0.002619948],"category_scores_gemma":[0.8792294,0.001750841,0.003239196,0.008855718,0.02007603,0.01066427,0.01139893,0.005461419,0.001026715],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004891441,"about_ca_system_score_gemma":0.005183619,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00180553,"about_ca_topic_score_gemma":0.001863044,"domain_scores_codex":[0.2660926,0.5193769,0.06144322,0.04364005,0.1070568,0.002390423],"domain_scores_gemma":[0.03221424,0.8001816,0.03614707,0.103712,0.02684278,0.0009023547],"domain_codex":"methods","domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.003534528,0.0005050948,0.2064272,0.01683554,0.01279786,0.001555511,0.07247434,0.01113032,0.006307982,0.1407044,0.03080142,0.4969258],"study_design_scores_gemma":[0.0009503124,0.001053686,0.1446095,0.01601884,0.004067644,0.002833635,0.01259789,0.0252607,0.02486308,0.6349392,0.1318739,0.0009316446],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3148224,0.05308382,0.487966,0.08299384,0.006691295,0.002654075,0.004265961,0.002661117,0.04486143],"genre_scores_gemma":[0.9029801,0.003338563,0.07670691,0.009177021,0.001807487,0.001913457,0.001646637,0.001162835,0.001266972],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.4059638,"threshold_uncertainty_score":0.5006257,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03935335336173296,"score_gpt":0.3372124674052018,"score_spread":0.2978591140434689,"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."}}