VisualAgentBench: Towards Large Multimodal Models as Visual Foundation Agents Paper • 2408.06327 • Published Aug 12 • 16