Reproduce “CineMME: From Dialogue Grounding to Narrative Reasoning in Multimodal Video Understanding” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jin Liu, Dawei Du, Yexiang Liu, Sijie Zhu, Fan Chen, Junxian Duan, Huaibo Huang, Ran He and 2 more
CineMME evaluates dialogue grounding and video reasoning on short English-language narrative clips. It measures speech timing, transcription, and speaker-face localization separately, then tests questions using dialogue, visual, and acoustic evidence. Gemini-3 Pro achieves 68.34% temporal overlap, 28.99% word error rate, and 7.74% face-track overlap; its 67.66% multiple-choice accuracy covers extractable answers and does not establish correct grounding.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.