{"id":"W2887879157","doi":"10.18653/v1/w18-5434","title":"Learning Explanations from Language Data","year":2018,"lang":"en","type":"preprint","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Banting and Best Diabetes Centre, University of Toronto; Bundesministerium für Bildung und Forschung","keywords":"Computer science; Linguistics; Psychology; Natural language processing; Cognitive psychology; Data science; Cognitive science; Philosophy","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["open_science","insufficient_payload"],"consensus_categories":["open_science"],"category_scores_codex":[0.0004612451,0.0002057354,0.0002063919,0.000139013,0.000185134,0.0006458604,0.005819956,0.000184025,0.000709374],"category_scores_gemma":[0.0003933991,0.0002011277,0.00004761233,0.0002073611,0.00005780421,0.0006697556,0.01006295,0.0005807799,0.002197513],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005407848,"about_ca_system_score_gemma":0.000175735,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004903203,"about_ca_topic_score_gemma":0.00105633,"domain_scores_codex":[0.9978251,0.0001281019,0.0003076341,0.001086055,0.0003506551,0.0003024331],"domain_scores_gemma":[0.9958355,0.000242363,0.0001569429,0.003533577,0.0001320077,0.0000996466],"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":[0.00002048949,0.0005709307,0.00321911,0.000147906,0.0006226657,0.0006140906,0.07404809,0.01323568,0.004288596,0.2134241,0.2772614,0.4125469],"study_design_scores_gemma":[0.00005107442,0.00004305406,0.000230305,0.00009295748,0.0000199168,0.000003004535,0.001953141,0.9204583,0.01326514,0.0344864,0.02877786,0.0006188941],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01425376,0.0002020399,0.9500796,0.001187093,0.001219895,0.0001891034,0.00006627836,0.0007276362,0.03207455],"genre_scores_gemma":[0.6552707,0.00008952864,0.3333425,0.0006890894,0.001262793,0.00004221853,0.001624984,0.00004159403,0.007636646],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9072226,"threshold_uncertainty_score":0.999559,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09718035733699655,"score_gpt":0.3463604548073858,"score_spread":0.2491800974703892,"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."}}