{"id":"W4389671215","doi":"10.1016/j.neucom.2023.127128","title":"Zero-shot multitask intent and emotion prediction from multimodal data: A benchmark study","year":2023,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Emotion and Mood Recognition","field":"Psychology","cited_by":13,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"Japan Society for the Promotion of Science","keywords":"Computer science; Modalities; Empathy; Realm; Artificial intelligence; Benchmark (surveying); Multimodal learning; Gesture; Multimodality; Human–computer interaction; Psychology; Social psychology","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":[0.0003680388,0.0001485604,0.000154505,0.0001640335,0.0001967147,0.0000573957,0.0001644004,0.00008059862,0.0001045596],"category_scores_gemma":[0.0001005216,0.0001502697,0.00002984849,0.0002533947,0.00002785687,0.0001573052,0.0003005769,0.0002562887,0.0002710393],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001540976,"about_ca_system_score_gemma":0.000007239478,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005100281,"about_ca_topic_score_gemma":0.00003342542,"domain_scores_codex":[0.9983264,0.0002468605,0.0003138519,0.000686997,0.000178444,0.0002475066],"domain_scores_gemma":[0.9992107,0.0001555458,0.0001054623,0.000394269,0.00005169608,0.00008233664],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0001818337,0.00184141,0.2614418,0.00004449234,0.0003027119,0.0002889229,0.02074368,0.0003809173,0.00752385,0.0000587692,0.01817148,0.6890202],"study_design_scores_gemma":[0.001894512,0.0002681583,0.9257144,0.00003948621,0.0000544192,0.00002258917,0.003567893,0.06758701,0.00002664224,0.00008720861,0.0005962773,0.0001413739],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9932707,0.00001895938,0.002029391,0.0001225135,0.002383672,0.0006039187,0.0001046081,0.0004614271,0.001004806],"genre_scores_gemma":[0.9981649,0.000008048906,0.0003141758,0.0001642001,0.0004202424,0.00001667271,0.0007551328,0.00002564659,0.0001310076],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6888788,"threshold_uncertainty_score":0.612782,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1071258844311998,"score_gpt":0.3521571728548367,"score_spread":0.2450312884236369,"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."}}