{"id":"W4226025707","doi":"10.1016/j.inffus.2022.03.001","title":"Multimodal Attentive Fusion Network for audio-visual event recognition","year":2022,"lang":"en","type":"article","venue":"Information Fusion","topic":"Music and Audio Processing","field":"Computer Science","cited_by":45,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Sherbrooke","funders":"European Regional Development Fund; Fonds de recherche du Québec – Nature et technologies; Nvidia","keywords":"Computer science; Audio visual; Event (particle physics); Modality (human–computer interaction); Window (computing); Fuse (electrical); Modalities; Artificial intelligence; Focus (optics); Multimodality; Code (set theory); Visualization; Speech recognition; Set (abstract data type); Multimedia","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0007119458,0.0008886187,0.0005921961,0.0006672315,0.0004351746,0.0005815612,0.0009913038,0.0007654502,0.004510932],"category_scores_gemma":[0.0009391105,0.0002409317,0.0006044582,0.0006977687,0.0002963801,0.001052658,0.001358561,0.0009197071,0.001270219],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004618171,"about_ca_system_score_gemma":0.0005477515,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003310251,"about_ca_topic_score_gemma":0.006147757,"domain_scores_codex":[0.9997153,0.00004303495,0.0000137041,0.00009932263,0.00006506089,0.00006357466],"domain_scores_gemma":[0.9997619,0.00007335545,0.00001921601,0.00002877197,0.00009382555,0.00002302895],"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":[0.000835304,0.0004114855,0.001906888,0.0001345058,0.0001593537,0.0002166563,0.0001400715,0.02863404,0.1137672,0.00384179,0.007492071,0.8424607],"study_design_scores_gemma":[0.0000284594,0.000419804,0.004985789,0.00003482306,0.0002274791,0.0002439022,0.00009577032,0.9094929,0.06991737,0.008678867,0.005822116,0.00005261843],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05261958,0.00170377,0.9370257,0.0002774796,0.0003063403,0.0001220205,0.0003915732,0.002250236,0.005303331],"genre_scores_gemma":[0.7673924,0.001182669,0.2169233,0.0005338704,0.0003475533,0.0001989471,0.001226774,0.0001206678,0.01207383],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004510932,"threshold_uncertainty_score":0.01509058,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01670245718060828,"score_gpt":0.2551667884741185,"score_spread":0.2384643312935102,"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."}}