{"id":"W6929255951","doi":"10.48448/m947-8n53","title":"Test-Time Self-Adaptive Small Language Models for Question Answering","year":2023,"lang":"en","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Robustness (evolution); Question answering; Language model; Benchmark (surveying); Transfer of learning; Memorization; Code (set theory); Noise (video)","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":[0.001889349,0.001359789,0.0008691943,0.0005765238,0.0003797647,0.0008684307,0.002750931,0.001578929,0.007381119],"category_scores_gemma":[0.01126926,0.0005460139,0.0008440515,0.0005576459,0.0006447254,0.002251838,0.00198879,0.002457596,0.004987717],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007843826,"about_ca_system_score_gemma":0.001158779,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004737635,"about_ca_topic_score_gemma":0.00914529,"domain_scores_codex":[0.9990572,0.0003333053,0.0000495416,0.0003428215,0.0001419138,0.00007519138],"domain_scores_gemma":[0.9966994,0.001962746,0.0001039127,0.0006639584,0.0004122012,0.0001577754],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001108327,0.000694152,0.004307965,0.0005358817,0.0002433574,0.0002281582,0.0004364645,0.2166588,0.02791748,0.004886996,0.04499366,0.6979887],"study_design_scores_gemma":[0.00007644585,0.0001049529,0.0003392755,0.00001313838,0.00001869994,0.00004520114,0.00003243797,0.985598,0.006721604,0.004319857,0.002711767,0.00001857561],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07648216,0.001792223,0.8082287,0.0008650678,0.0004535107,0.0004207994,0.002672004,0.1045958,0.004489651],"genre_scores_gemma":[0.5409953,0.0003465565,0.4356791,0.001053845,0.0001478703,0.0009671872,0.009852157,0.003268074,0.00768991],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007381119,"threshold_uncertainty_score":0.0246923,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04519374099210151,"score_gpt":0.31235550191181,"score_spread":0.2671617609197085,"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."}}