arXivAmrita Mazumdar, Seonwook Park, Rajarshi Roy, Nikhil Srihari, Shengze Wang, Yuhao Zhou, Julia Wang, Koki Nagano, Shalini De MelloThu, May 28, 2026, 10:20 AM PDT
score 14.8
First benchmark for conversational AI that sees and speaks simultaneously
Original: VideoFDB: Evaluating Full-Duplex Vision-Speech Capabilities in Conversational Agents
Source: arxiv.org ↗
Writing ELI5 summary…