Can Agents Read the Room? Benchmarking Visual Social Intelligence in Multimodal Simulation

In three linesAgentViSS benchmark evaluates visual social intelligence of multimodal agents in social simulations. 240 scenarios, 585 roles, 2,340 instances test whether MLLMs use visual cues (expressions, posture, gaze) to guide interactions. Seven models evaluated show gap: expression and conflict handling near saturation, interaction regulation and visually grounded outcomes remain substantially harder.

Summary generated by Claude — human-verified