Signal2026-07-20
Papers With Code

VideoRAE: Taming Video Foundation Models for Generative Modeling via Representation Autoencoders

Part of

Unified World Modeling In Video Generation

Open primary sourceSee the whole picture