# TRIAGE: VLA inference acceleration requires both reusing and pruning visual tokens

- Source: arXiv cs.RO
- Published: 2026-10-27T04:00:00.000Z
- RoboHot: https://robohot.news/en/items/o9inn8zb9xdbrpbsq6vfkxv2z
- Original: https://arxiv.org/abs/2610.34519

## Summary

TRIAGE proposes a training-free visual token acceleration method that decides whether to reuse cached KV or prune based on a token drift threshold, and fixes the recomputation budget to directly control prefill latency.