VidLaDA: Bidirectional Diffusion Large Language Models for Efficient Video Understanding

The repo is coming soon.