{"id":"W4386256573","doi":"10.3390/stats6030056","title":"Investigating Self-Rationalizing Models for Commonsense Reasoning","year":2023,"lang":"en","type":"article","venue":"Stats","topic":"Topic Modeling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal","funders":"Alliance de recherche numérique du Canada; Natural Sciences and Engineering Research Council of Canada; Government of Canada","keywords":"Computer science; Generative grammar; Leverage (statistics); Commonsense reasoning; Language model; Transformer; Artificial intelligence; Natural language processing; Natural language understanding; Generative model; Natural language; Representation (politics)","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0004544136,0.00007732608,0.00008723541,0.00007412752,0.0002251374,0.0001144682,0.0002324456,0.00002968803,7.970669e-7],"category_scores_gemma":[0.0001215614,0.00008225307,0.00003402931,0.0002849319,0.00001155142,0.0003922604,0.0001518409,0.00006695999,0.0000174407],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003136059,"about_ca_system_score_gemma":0.00009125085,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002468743,"about_ca_topic_score_gemma":0.00001002869,"domain_scores_codex":[0.9990878,0.00003687046,0.0001714678,0.0002625873,0.000185246,0.0002560453],"domain_scores_gemma":[0.9992555,0.000257924,0.00005730741,0.0002897593,0.00007019073,0.00006926439],"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":[0.000001016511,0.000009723216,0.0003199418,0.00004659605,0.00001879384,0.000009741979,0.00856918,0.1335933,0.0007554715,0.84258,0.004362532,0.009733771],"study_design_scores_gemma":[0.0001184658,0.000008833933,0.00007266493,0.00002682011,0.00000187088,0.000003599234,0.00006902479,0.8603777,0.0002015575,0.1384586,0.0005765994,0.00008429791],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06891495,0.00002944089,0.927686,0.001257173,0.0002026444,0.0001614523,0.000004301157,0.000737198,0.0010068],"genre_scores_gemma":[0.4431973,0.000002920438,0.5561799,0.000367753,0.00007166979,0.00003470184,0.000008615038,0.00001057327,0.0001265823],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7267844,"threshold_uncertainty_score":0.3354183,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09268685221712009,"score_gpt":0.3107939534302409,"score_spread":0.2181071012131208,"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."}}