{"id":"W4405033244","doi":"10.1109/iccv51701.2025.01364","title":"FLOAT: Generative Motion Latent Flow Matching for Audio-Driven Talking Portrait","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Music and Audio Processing","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Portrait; Generative grammar; Matching (statistics); Float (project management); Flow (mathematics); Motion (physics); Computer science; Generative model; Acoustics; Speech recognition; Computer vision; Artificial intelligence; Art; Visual arts; Engineering; Marine engineering; Mathematics; Geometry; Physics","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":[0.0003697254,0.0003494125,0.000417558,0.0001826707,0.0003369421,0.0006284826,0.0009979289,0.0002555029,0.00003805786],"category_scores_gemma":[0.0000274777,0.0003191382,0.0002548155,0.0001716371,0.00002617679,0.0003456491,0.001504145,0.0004153161,0.000008366371],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001344424,"about_ca_system_score_gemma":0.0003834898,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004565254,"about_ca_topic_score_gemma":0.00002536396,"domain_scores_codex":[0.9978184,0.00005335619,0.0004587825,0.0009839524,0.0002985597,0.000386948],"domain_scores_gemma":[0.9987822,0.00007053839,0.0002915627,0.0005635417,0.000207969,0.00008416843],"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":[0.00001074542,0.0001232272,0.0001042716,0.001374593,0.0002747952,0.0000295733,0.006479457,0.317117,0.001975043,0.07643799,0.01337393,0.5826994],"study_design_scores_gemma":[0.0002790719,0.0000189119,0.0001800672,0.0006087285,0.00003740998,0.000006242066,0.00002673327,0.9363126,0.005285325,0.05509807,0.001647679,0.0004991738],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00112299,0.0001668055,0.9862701,0.002852099,0.001939401,0.0005825261,0.00002806245,0.0003742142,0.006663827],"genre_scores_gemma":[0.07699492,0.00002236036,0.9152836,0.002039852,0.0005328808,0.0001613859,0.00007907784,0.00001795121,0.004868018],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.6191956,"threshold_uncertainty_score":0.9999261,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02837923795009671,"score_gpt":0.2749640554177935,"score_spread":0.2465848174676967,"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."}}