WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction Paper • 2605.29341 • Published 5 days ago • 10
CiteVQA: Benchmarking Evidence Attribution for Trustworthy Document Intelligence Paper • 2605.12882 • Published 20 days ago • 269
SocialOmni: Benchmarking Audio-Visual Social Interactivity in Omni Models Paper • 2603.16859 • Published Mar 17 • 248