Reroute, Don't Remove: Recoverable Visual Token Routing for Vision-Language Models

Cheng-Yu Yang, Shao-Yuan Lo, Yu-Lun Liu

Thursday at 04:00

1 Views

0 Comments

arXiv:2606.12412v1 Announce Type: cross Abstract: Vision-language models (VLMs) project images into hundreds to thousands of visual tokens, making decoder inference expensive in both attention computation and KV-cache memory. Existing visual-token reduction methods largely follow a rank-and-remove paradigm: they score visual tokens, keep a...

Read the full article at the source.

Read Original Article

Was this helpful?