{"id":"W4416034471","doi":"10.18653/v1/2025.findings-emnlp.1340","title":"QEVA: A Reference-Free Evaluation Metric for Narrative Video Summarization with Multimodal Question Answering","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":"Artificial Intelligence in Medicine (Canada)","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science and ICT, South Korea; Chung-Ang University","keywords":"Automatic summarization; Question answering; Metric (unit); Narrative; 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":[0.002742718,0.0005564601,0.0005129286,0.00120967,0.001158226,0.0007060734,0.00150603,0.0003272751,0.000120224],"category_scores_gemma":[0.003158847,0.0005159281,0.000122773,0.004618055,0.0001530162,0.00120572,0.000470159,0.0005450112,0.00002901031],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007909323,"about_ca_system_score_gemma":0.001232906,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003065244,"about_ca_topic_score_gemma":0.0008290447,"domain_scores_codex":[0.995055,0.0007121003,0.0009045315,0.001647037,0.001046814,0.0006345014],"domain_scores_gemma":[0.9936374,0.0009983841,0.0005812068,0.001797925,0.002813364,0.0001717577],"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":[0.000375711,0.0005874897,0.009887978,0.0002907104,0.000313445,8.176897e-7,0.00624948,0.1843026,0.001823919,0.2513966,0.000759388,0.5440119],"study_design_scores_gemma":[0.003432791,0.000474839,0.06028695,0.0002964964,0.00025735,0.000002514052,0.000321421,0.9211147,0.00100559,0.01159371,0.0006993117,0.0005142554],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03131613,0.0002651104,0.9491794,0.006105961,0.0004542761,0.004743724,0.00002253268,0.0003821148,0.00753081],"genre_scores_gemma":[0.8048521,0.00002323218,0.191888,0.0002158844,0.00008902933,0.001644519,0.0001447074,0.00003053345,0.001112019],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.773536,"threshold_uncertainty_score":0.9997292,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02417721802176372,"score_gpt":0.3427726956916791,"score_spread":0.3185954776699154,"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."}}