WorksComputer Science
CineMME: From Dialogue Grounding to Narrative Reasoning in Multimodal Video Understanding
Abstract
Narrative video understanding requires more than recognizing visible events: models must extract evidence from dialogue, bind it to speakers, and use it to infer narrative meaning. Existing video benchmarks evaluate aspects of this problem, including video question answering, audio-visual reasoning, and temporal grounding, but rarely test whether these abilities form a coherent dialogue-to-plot evidence chain within the same clips. We introduce CineMME, a human-verified benchmark designed to diagnose dialogue-grounded plot understanding, supported by approximately expert person-hours of annotation. CineMME contains curated short narrative video clips and two linked tracks: CineMME-Grounding, with timestamped speech segments and speaker face boxes for temporal dialogue localization, transcription, and active-speaker grounding; and CineMME-Reasoning, with multiple-choice and open-ended questions spanning perception, audio, narrative, social, and scene-presentation understanding. Our main evaluation compares MLLMs on reasoning and omni-modal models on grounding. The results reveal a pronounced who-what-when decoupling: Gemini-3 Pro localizes dialogue in time ( tIoU) and transcribes it with a WER of , yet achieves only vIoU for active-speaker grounding while achieving MCQ accuracy on questions with extractable answers. Modality ablations further show that vision is not uniformly beneficial: it improves performance on perception-heavy questions but can hurt dialogue-centric reasoning. CineMME provides a compact diagnostic testbed for evaluating how multimodal video models connect dialogue, speakers, visual evidence, and narrative meaning.