Abstract: Large multimodal models (LMMs) with advanced video analysis capabilities have recently garnered significant attention. However, most evaluations rely on traditional methods like ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results