{"id":"W7103235250","doi":"","title":"Data-Efficient Learning via Clustering-Based Sensitivity Sampling: Foundation Models and Beyond","year":2024,"lang":"en","type":"article","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Google (Canada)","funders":"Deutsche Forschungsgemeinschaft; Austrian Science Fund; European Commission; Institute of Science and Technology Austria","keywords":"Sensitivity (control systems); Foundation (evidence); Relation (database); Noise (video); Field (mathematics)","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":[0.008323007,0.0001951945,0.0001912237,0.0001657519,0.0003940161,0.0009635593,0.0007124284,0.00008078224,0.00001010052],"category_scores_gemma":[0.0004002374,0.0001961666,0.00005883952,0.0004707934,0.00011529,0.0005676738,0.000949326,0.0003320485,0.00001488186],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005369964,"about_ca_system_score_gemma":0.0001523795,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001321755,"about_ca_topic_score_gemma":0.0003203197,"domain_scores_codex":[0.9951504,0.003112811,0.0002826331,0.0008591251,0.0003109161,0.0002840677],"domain_scores_gemma":[0.9961129,0.001635791,0.0001136551,0.001528711,0.0004616941,0.0001471777],"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":[0.000004386662,0.0001359426,0.00002481468,0.0000996953,0.00002591962,0.00001052652,0.003190504,0.01038444,0.003917823,0.4011479,0.0001011803,0.5809569],"study_design_scores_gemma":[0.0001509625,4.819083e-7,0.0000734691,0.0002279194,0.00001396195,0.00002778213,0.00001007935,0.9815631,0.002104282,0.01009135,0.005532935,0.0002036768],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002379996,0.0005161545,0.9860809,0.004890259,0.0002299748,0.0001874231,0.00001189598,0.0004109611,0.005292458],"genre_scores_gemma":[0.4887967,0.00002578792,0.5106113,0.0001112641,0.00001595495,0.000006867779,0.0000770093,0.00001649969,0.0003386108],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9711787,"threshold_uncertainty_score":0.9291624,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03772344390659249,"score_gpt":0.2774088458968559,"score_spread":0.2396854019902634,"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."}}