TOMATO: Assessing Visual Temporal Reasoning Capabilities in Multimodal Foundation Models Paper • 2410.23266 • Published Oct 30 • 20