{"id":"W2973802306","doi":"10.1109/iccv.2019.00900","title":"Watch, Listen and Tell: Multi-Modal Weakly Supervised Dense Event Captioning","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; Canadian Institute for Advanced Research; University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Closed captioning; Computer science; Event (particle physics); Modal; Focus (optics); Feature (linguistics); Representation (politics); Modalities; Ranging; Speech recognition; Audio visual; Natural language processing; Artificial intelligence; Multimedia; Image (mathematics); Linguistics","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004499485,0.0004083632,0.0004159039,0.0001388249,0.000201599,0.0005214847,0.001463348,0.0003186973,0.00005764828],"category_scores_gemma":[0.00009312705,0.0003819989,0.0001549002,0.0001619382,0.0000635134,0.0001947968,0.003347923,0.001013133,0.0003938691],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001176272,"about_ca_system_score_gemma":0.0001819906,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005102813,"about_ca_topic_score_gemma":0.0001150154,"domain_scores_codex":[0.9973139,0.0001645545,0.0004567916,0.00127775,0.0003778089,0.0004092473],"domain_scores_gemma":[0.9974617,0.0001978248,0.0002085979,0.001745195,0.000175084,0.0002116499],"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":[0.00005190748,0.001385408,0.2373478,0.001570098,0.0006352888,0.00006637519,0.01711651,0.4654018,0.01355373,0.08461976,0.002848421,0.1754028],"study_design_scores_gemma":[0.0004025211,0.00002524692,0.09148768,0.00007846658,0.00002055256,0.00002045826,0.0000268506,0.9056031,0.0001560023,0.0006318573,0.001075275,0.0004720409],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2814127,0.0001412651,0.7107077,0.004273359,0.0004170821,0.000843416,0.000009789968,0.0005949383,0.001599714],"genre_scores_gemma":[0.7563062,0.00003239735,0.2399066,0.0002389982,0.00007824974,0.0001164821,0.00003371497,0.00003257192,0.003254757],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.4748935,"threshold_uncertainty_score":0.9998632,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02367218328816387,"score_gpt":0.2905838013795247,"score_spread":0.2669116180913608,"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."}}