{"id":"W3211673252","doi":"10.5281/zenodo.3982680","title":"Related codes for \"Why Do Masked Neural Language Models Still Need Commonsense Knowledge to Handle Semantic Variations in Question Answering?\"","year":2020,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Question answering; Commonsense knowledge; Computer science; Natural language processing; Artificial intelligence; Commonsense reasoning; Knowledge-based systems","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003127754,0.001037814,0.0008657834,0.001602628,0.002050671,0.003559012,0.002251661,0.004643602,0.09455059],"category_scores_gemma":[0.03920053,0.000512568,0.001574811,0.001500343,0.002516499,0.006667472,0.002904583,0.003678628,0.03150907],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003271628,"about_ca_system_score_gemma":0.002204703,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005278619,"about_ca_topic_score_gemma":0.002583905,"domain_scores_codex":[0.9963562,0.0008376747,0.0002922831,0.0008228197,0.001363073,0.0003279186],"domain_scores_gemma":[0.9782087,0.007711877,0.001027151,0.003550599,0.008900271,0.0006014915],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001806465,0.00007330113,0.0005047124,0.0002863865,0.00003696869,0.0001619192,0.0002663263,0.001053859,0.002250972,0.3355287,0.6248922,0.03476403],"study_design_scores_gemma":[0.0001088719,0.00005901079,0.001922807,0.0001702392,0.0000417385,0.0004766025,0.0002632417,0.03119458,0.009370142,0.5482721,0.4079707,0.0001499809],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"dataset","genre_scores_codex":[0.008951596,0.004457198,0.3573771,0.2902876,0.1841348,0.0004394263,0.01141329,0.005390952,0.1375481],"genre_scores_gemma":[0.3709563,0.004949335,0.1291098,0.08954392,0.07715189,0.001288309,0.01954912,0.008224454,0.2992268],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.09455059,"threshold_uncertainty_score":0.3163032,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04508286160316482,"score_gpt":0.2663745417429629,"score_spread":0.2212916801397981,"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."}}