{"id":"W4415955467","doi":"10.1109/iccv51701.2025.00074","title":"MultiVerse: A Multi-Turn Conversation Benchmark for Evaluating Large Vision and Language Models","year":2025,"lang":"","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Ministry of Science and ICT, South Korea","keywords":"Conversation; Benchmark (surveying); Set (abstract data type); Perception; Context (archaeology); Key (lock)","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":[0.001594953,0.000340925,0.0003382522,0.0002937173,0.0007260903,0.0003901806,0.0005584304,0.0002042774,0.00007988051],"category_scores_gemma":[0.0005702006,0.0003348115,0.0001291263,0.0005891669,0.00009194624,0.0008254114,0.000663058,0.0003340603,0.00002436794],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001431523,"about_ca_system_score_gemma":0.0001971476,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009506146,"about_ca_topic_score_gemma":0.00010813,"domain_scores_codex":[0.9972634,0.0002632153,0.000524512,0.001101336,0.000328178,0.0005193568],"domain_scores_gemma":[0.997579,0.0008880125,0.0002663055,0.0007244046,0.0004013988,0.0001408826],"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":[0.0001608623,0.001035176,0.004741739,0.0006686631,0.000226616,0.000002847052,0.05198038,0.03964943,0.02005292,0.1756955,0.0006718193,0.705114],"study_design_scores_gemma":[0.004115304,0.0001481208,0.01590785,0.0001510957,0.00007566044,0.000001256218,0.001319551,0.9759289,0.0003909597,0.001449214,0.0002032147,0.0003088658],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08133004,0.0005452146,0.9117511,0.002703636,0.0002871189,0.002076652,0.0000517533,0.0001700581,0.001084408],"genre_scores_gemma":[0.7153011,0.00002183269,0.2826985,0.000453004,0.00002700003,0.0001519615,0.00003128481,0.00001346767,0.001301788],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9362795,"threshold_uncertainty_score":0.9999104,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02903482643643528,"score_gpt":0.3890024476712032,"score_spread":0.3599676212347679,"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."}}