V-LynX: Token Interface Alignment for Video+X LLMs 事件

V-LynX: Token Interface Alignment for Video+X LLMs arXiv:2606.00508v1 Announce Type: new Abstract: This study introduces an intriguing phenomenon in Video LLMs: rather than merely translating frames into textual embeddings, Video LLMs establish a continuous manifold, token interface, allowing visual tokens to operate as standalone entities within the architecture. Exploiting this discovery, we propose V-LynX, a scalable framework that integrates novel modalities into Video LLMs by repurposing t