{"id":"W4285263440","doi":"10.18653/v1/2022.acl-long.129","title":"Probing as Quantifying Inductive Bias","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","topic":"Topic Modeling","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Max Planck ETH Center for Learning Systems; McGill University; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; National Science Foundation","keywords":"Inductive bias; Computer science; ENCODE; Variety (cybernetics); Task (project management); Bayesian probability; Artificial intelligence; Natural language processing; Range (aeronautics); Position paper; Multi-task learning; Machine learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02486958,0.001310505,0.001469648,0.002431373,0.001521464,0.003190638,0.00273847,0.002887548,0.003314134],"category_scores_gemma":[0.1587383,0.001068361,0.001035624,0.002782855,0.006869228,0.01299411,0.008224417,0.005480501,0.0006755653],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002293624,"about_ca_system_score_gemma":0.001949654,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001278104,"about_ca_topic_score_gemma":0.001231328,"domain_scores_codex":[0.9825093,0.01066288,0.0006788668,0.002352877,0.003086391,0.0007098279],"domain_scores_gemma":[0.8467169,0.1294807,0.006136586,0.01292935,0.003613087,0.001123461],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007434618,0.0001877825,0.01569947,0.0006290749,0.0002661052,0.0001876946,0.00236576,0.1225481,0.008044601,0.6503749,0.00367478,0.1952783],"study_design_scores_gemma":[0.00004312461,0.0001067211,0.001784942,0.00008017493,0.00006201019,0.0001465661,0.0001717632,0.1683383,0.005184087,0.8201913,0.003835346,0.00005561113],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03904122,0.0007167027,0.9510226,0.002190924,0.00007060238,0.0001104168,0.0002641875,0.0005803875,0.006003082],"genre_scores_gemma":[0.7392645,0.0006603995,0.2547292,0.001409282,0.0003220974,0.0006217146,0.0006805407,0.0004310148,0.001881222],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02486958,"threshold_uncertainty_score":0.1315244,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02919653589517536,"score_gpt":0.2628133028168727,"score_spread":0.2336167669216974,"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."}}