{"id":"W3207166518","doi":"10.18653/v1/2022.naacl-main.341","title":"Symbolic Knowledge Distillation: from General Language Models to Commonsense Models","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies","topic":"Topic Modeling","field":"Computer Science","cited_by":150,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Naval Information Warfare Center Pacific; Natural Sciences and Engineering Research Council of Canada; Defense Advanced Research Projects Agency; Allen Institute for Artificial Intelligence","keywords":"Computer science; Computational linguistics; Language model; Artificial intelligence; Natural language processing; Cognitive science; Linguistics; Philosophy; Psychology","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":[0.001998912,0.001046646,0.001428163,0.001649417,0.00115735,0.003559233,0.003037818,0.001443379,0.009483786],"category_scores_gemma":[0.01358358,0.0008982492,0.001680283,0.002142123,0.0021915,0.009962929,0.004991059,0.004271008,0.00209305],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001362651,"about_ca_system_score_gemma":0.001957456,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00835254,"about_ca_topic_score_gemma":0.01538977,"domain_scores_codex":[0.9987614,0.0005817107,0.00007014883,0.0002221642,0.0002592082,0.0001053669],"domain_scores_gemma":[0.9930997,0.005292176,0.0001844574,0.0008023737,0.0004468377,0.0001744807],"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":[0.0004615095,0.0001867368,0.0007324966,0.0004113901,0.0001539163,0.0002019563,0.0005431865,0.3058087,0.0009029977,0.3870051,0.01460658,0.2889853],"study_design_scores_gemma":[0.00002392216,0.0000125771,0.00004015896,0.00003352448,0.0000154871,0.00001635209,0.00004174657,0.6219463,0.0004905505,0.3751429,0.00222351,0.00001292855],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0147803,0.001330773,0.9710153,0.002328981,0.0001602912,0.00006852211,0.0008793494,0.002405458,0.00703109],"genre_scores_gemma":[0.5396805,0.001669056,0.4452918,0.0007072332,0.0003063484,0.0003551092,0.003119287,0.0009610816,0.00790944],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009483786,"threshold_uncertainty_score":0.03172642,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02968002282188094,"score_gpt":0.2597489079193148,"score_spread":0.2300688850974339,"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."}}