{"id":"W3196766446","doi":"10.48550/arxiv.2109.02613","title":"Class Semantics-based Attention for Action Detection","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Human Pose and Action Recognition","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Feature (linguistics); Action (physics); Encoder; Semantics (computer science); Class (philosophy); Artificial intelligence; Task (project management); Mechanism (biology); Pattern recognition (psychology); Attention network","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001736125,0.0002125627,0.0002033689,0.0002973109,0.0002651786,0.0002631884,0.0004104856,0.0003101763,0.00002703816],"category_scores_gemma":[0.00002354417,0.0002770284,0.0003233216,0.0003457523,0.00003097725,0.0005542151,0.0002351339,0.0003438432,0.00004159401],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002534188,"about_ca_system_score_gemma":0.0001428457,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003299953,"about_ca_topic_score_gemma":0.0001771407,"domain_scores_codex":[0.9985518,0.0001098732,0.0001726138,0.0008560347,0.00008260908,0.0002270096],"domain_scores_gemma":[0.9986567,0.00007016544,0.0002674077,0.000575998,0.0003447374,0.00008502381],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005842284,0.002213283,0.002324657,0.003627802,0.001189589,0.0006194976,0.0006362674,0.7083112,0.05596485,0.08258904,0.001651711,0.1402879],"study_design_scores_gemma":[0.0006906981,0.0000774043,0.0009462946,0.0001190753,0.0001138345,0.000004189179,0.00010862,0.9758185,0.01336359,0.007547706,0.0008309604,0.0003790944],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3306972,0.00000610791,0.6674418,0.00008790007,0.001113716,0.0002483469,0.000006126985,0.0002145065,0.0001842089],"genre_scores_gemma":[0.9973902,0.00003446632,0.001487544,0.0001120448,0.0001595937,0.000005477171,0.000115064,0.00001573411,0.0006799466],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6666929,"threshold_uncertainty_score":0.9999682,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1187528620764375,"score_gpt":0.2121582252449545,"score_spread":0.09340536316851697,"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."}}