
TempoFit: Plug-and-Play Layer-Wise Temporal KV Memory for Long-Horizon Vision-Language-Action Manipulation
IROS 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems
A training-free retrofit that gives frozen VLA policies history awareness by caching intermediate-layer prefix K/V states — no parameter change, no longer inputs. +4.0 points on LIBERO-Long with π0.5, and up to +14.2% full-task success on a real Realman RM-65B.
2026