{"id":"W3095478416","doi":"10.1109/cvprw53098.2021.00181","title":"An Improved Attention for Visual Question Answering","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; Canadian Institute for Advanced Research; University of British Columbia","funders":"","keywords":"Question answering; Computer science; Benchmark (surveying); Artificial intelligence; Encoder; Context (archaeology); Task (project management); Natural language; Modality (human–computer interaction); Relation (database); Information retrieval; Natural language processing; Data mining","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":[0.001085699,0.001342375,0.001101434,0.001903534,0.000589016,0.0009328924,0.002273731,0.002082732,0.00608343],"category_scores_gemma":[0.00272148,0.000395157,0.001513489,0.001220061,0.0007085847,0.002520978,0.00229832,0.001815443,0.001493083],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00142853,"about_ca_system_score_gemma":0.001189344,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01740806,"about_ca_topic_score_gemma":0.01321642,"domain_scores_codex":[0.9989318,0.0002355964,0.0000486716,0.0004462761,0.0001907525,0.0001468509],"domain_scores_gemma":[0.9992952,0.0002931813,0.00003285366,0.0001166999,0.0002057692,0.00005625878],"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":[0.0004107984,0.000267647,0.001439429,0.0004406129,0.0001623008,0.00029238,0.0004143224,0.0315058,0.04844276,0.01177318,0.02857918,0.8762715],"study_design_scores_gemma":[0.000101708,0.0002655714,0.001850168,0.00005109212,0.0001737875,0.0004018906,0.0001161319,0.9167144,0.02284509,0.03753226,0.01989994,0.00004805536],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02402758,0.003949417,0.9541308,0.001169442,0.000293051,0.0002721328,0.0009759177,0.009835903,0.005345788],"genre_scores_gemma":[0.5206565,0.001734242,0.4547778,0.002323124,0.0006715441,0.0004017307,0.005297131,0.0005715585,0.0135664],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01740806,"threshold_uncertainty_score":0.03461349,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01438235687974552,"score_gpt":0.349374374597222,"score_spread":0.3349920177174764,"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."}}