{"id":"W3113945743","doi":"10.18653/v1/2020.semeval-1.76","title":"TR at SemEval-2020 Task 4: Exploring the Limits of Language-model-based Common Sense Validation","year":2020,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Thomson Reuters (Canada)","funders":"","keywords":"Computer science; Commonsense knowledge; SemEval; Concreteness; Language model; Natural language processing; Commonsense reasoning; Task (project management); Artificial intelligence; Embedding; Knowledge graph; Language understanding; Common sense; Domain knowledge; Cognitive psychology; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002099575,0.0001012084,0.0001350633,0.00002392953,0.00008118154,0.00004607283,0.0004561689,0.00002652339,0.00001489417],"category_scores_gemma":[0.00005043382,0.00007370659,0.00006359543,0.0002228922,0.00001885539,0.0002405376,0.0002153267,0.00009378119,0.00003021493],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000288988,"about_ca_system_score_gemma":0.0000369978,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003426716,"about_ca_topic_score_gemma":0.00001521862,"domain_scores_codex":[0.9989622,0.00008332817,0.0002346232,0.0002742404,0.0002891911,0.0001564643],"domain_scores_gemma":[0.9991558,0.0001275359,0.00007923473,0.0005192466,0.0000443463,0.00007384596],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005033239,0.00007563346,0.001102407,0.000144183,0.00004100268,0.00004561379,0.03241215,0.7554331,0.1347046,0.01755839,0.00227723,0.05615536],"study_design_scores_gemma":[0.000188791,0.00002654072,0.00007180937,0.000009829333,0.000006176889,0.000001621727,0.00009832857,0.8853975,0.1138375,0.0001081729,0.000171826,0.00008191283],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4538403,0.00002450759,0.5399376,0.00506214,0.00007585472,0.00009566436,0.000001348316,0.0001157885,0.0008468009],"genre_scores_gemma":[0.9738564,0.000003720432,0.02417649,0.00176262,0.00006981518,0.00001102983,0.00000335819,0.000008351089,0.0001082453],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5200161,"threshold_uncertainty_score":0.3005667,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09066099485163504,"score_gpt":0.2827639114547911,"score_spread":0.1921029166031561,"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."}}