{"id":"W4407200740","doi":"10.1007/s10994-024-06659-z","title":"Robust query performance prediction for dense retrievers via adaptive disturbance generation","year":2025,"lang":"en","type":"article","venue":"Machine Learning","topic":"Advanced Image and Video Retrieval Techniques","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo; Toronto Metropolitan University","funders":"","keywords":"Disturbance (geology); Computer science; Artificial intelligence; Control theory (sociology); Pattern recognition (psychology); Biology; Control (management)","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":[0.0003333544,0.0001421316,0.0001438636,0.0001204569,0.0003963002,0.00009710026,0.0002796412,0.00006921197,0.000003715768],"category_scores_gemma":[0.0002097939,0.0001379222,0.00005989461,0.0004770903,0.00003049933,0.0008796548,0.0001145842,0.000262198,0.000002932203],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001132562,"about_ca_system_score_gemma":0.00004045882,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001483558,"about_ca_topic_score_gemma":0.000005036832,"domain_scores_codex":[0.9989788,0.00006067678,0.0002115196,0.0003836025,0.0001457228,0.0002197119],"domain_scores_gemma":[0.9993785,0.00009037515,0.0001088556,0.0002494773,0.0001381989,0.00003461133],"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":[0.0002118621,0.00006940781,0.02109823,0.0001095226,0.00005766243,0.000006624232,0.0004745922,0.03825052,0.01239474,0.01095448,0.00270195,0.9136704],"study_design_scores_gemma":[0.000280468,0.0002547927,0.003439075,0.00004614547,0.000009572662,0.000003944636,0.000006614224,0.9493942,0.03183642,0.0006829069,0.01390789,0.0001380014],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007208,0.0005049274,0.9904185,0.000247417,0.0002818778,0.0003003437,0.000004358506,0.0004732916,0.0005613058],"genre_scores_gemma":[0.824052,0.0001520625,0.1723106,0.0002718059,0.0001276173,0.00005645553,0.00002699358,0.00001255471,0.002989924],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9135324,"threshold_uncertainty_score":0.5624303,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02459554223764634,"score_gpt":0.2541639193749147,"score_spread":0.2295683771372684,"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."}}