VideoLLM-MoD: Efficient Video-Language Streaming with Mixture-of-Depths Vision Computation

Devs

VideoLLM-MoD: Efficient Video-Language Streaming with Mixture-of-Depths Vision Computation | Read Paper on Bytez