{"id":"W3106697459","doi":"10.18653/v1/2020.aacl-main.48","title":"Multimodal Pretraining for Dense Video Captioning","year":2020,"lang":"en","type":"preprint","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Closed captioning; Timeline; Computer science; Leverage (statistics); Variety (cybernetics); Multimedia; Construct (python library); Artificial intelligence; Human–computer interaction; Natural language processing; Image (mathematics)","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":[0.0003809892,0.0003334065,0.0003563816,0.000128946,0.0002603306,0.0004862454,0.001819476,0.0002698615,0.00003027127],"category_scores_gemma":[0.0005535504,0.0003502077,0.0002361641,0.0001990528,0.00003880547,0.0001636019,0.001872739,0.0009027802,0.0001166662],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009645816,"about_ca_system_score_gemma":0.0002400313,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006090592,"about_ca_topic_score_gemma":0.00001720371,"domain_scores_codex":[0.9975939,0.00008035083,0.0004265733,0.001228947,0.0002901024,0.0003801044],"domain_scores_gemma":[0.9977616,0.0004550763,0.0002741621,0.001118507,0.0001825949,0.0002080344],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004122889,0.0002001903,0.003275149,0.000783506,0.0003339214,0.00001597556,0.01331047,0.2408458,0.005038639,0.4706466,0.005446008,0.2600625],"study_design_scores_gemma":[0.0003088488,0.00003433163,0.003317076,0.00005283125,0.00001799945,0.000004939943,0.0000235242,0.9763381,0.0002293278,0.01790624,0.001376244,0.0003904703],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.007071671,0.00004257698,0.978328,0.008885398,0.0004285028,0.00111602,0.00002078401,0.001315374,0.00279161],"genre_scores_gemma":[0.4991922,0.000001657908,0.4995641,0.0004054213,0.000191591,0.000461865,0.00003950987,0.00002486724,0.0001187984],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7354923,"threshold_uncertainty_score":0.999895,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04438772172069969,"score_gpt":0.3200176612216676,"score_spread":0.2756299395009679,"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."}}