{"id":"W4385801313","doi":"10.1109/cvprw59228.2023.00496","title":"DeCAtt: Efficient Vision Transformers with Decorrelated Attention Heads","year":2023,"lang":"en","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Transformer; Overfitting; Artificial intelligence; Regularization (linguistics); Machine learning; Computer engineering; Engineering; Artificial neural network; Electrical engineering","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":[0.0001005508,0.0001036197,0.00008310364,0.0001147149,0.000178513,0.00004671274,0.000314923,0.00003551869,0.0000126223],"category_scores_gemma":[0.00000275241,0.00007759952,0.00003812193,0.001831137,0.0000339334,0.0002135611,0.00005488319,0.00008894147,0.0005863123],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003251816,"about_ca_system_score_gemma":0.00001911996,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004483928,"about_ca_topic_score_gemma":0.00001488023,"domain_scores_codex":[0.9989687,0.00001677608,0.0001515089,0.0003566099,0.0002268002,0.0002795909],"domain_scores_gemma":[0.9994504,0.00006751476,0.00003607343,0.0003107105,0.00004539975,0.00008986198],"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":[0.00004768914,0.0002601075,0.001659622,0.00002177092,0.00003717906,0.00003431486,0.0005105466,0.3442528,0.01631716,0.09805847,0.006814744,0.5319856],"study_design_scores_gemma":[0.0004202969,0.0001379279,0.01497679,0.00002095688,0.000005680513,0.00001900163,0.00002659304,0.9789118,0.001529904,0.0007374802,0.00301606,0.0001975604],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1389964,0.00001235707,0.8561606,0.002095866,0.00009390991,0.0002465066,5.283885e-7,0.0008453135,0.001548499],"genre_scores_gemma":[0.9783061,0.00002789539,0.02064801,0.0001691061,0.00001402092,0.00005121071,0.00001301645,0.00001216655,0.0007584948],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8393096,"threshold_uncertainty_score":0.7536054,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009890696637562682,"score_gpt":0.2666187037926013,"score_spread":0.2567280071550386,"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."}}