MM-Vet v2: A Challenging Benchmark to Evaluate Large Multimodal Models for Integrated Capabilities Paper โข 2408.00765 โข Published Aug 1 โข 12 โข 9