Reflex: Real-Time VLA Control Through Streaming Inference
Reflex is a framework that enables real-time streaming inference for flow matching Vision-Language-Action (VLA) models by leveraging a Timestep-Invariance Property. It partitions attention context into static, sliding, and dynamic regions, allowing O(1) incremental cache updates, and introduces AdaRMSNorm to prevent numerical instability. Reflex achieves a 2.58× inference speedup and stable 50Hz streaming on benchmarks, reducing reaction latency by up to 54%.
Why it matters: This work addresses a key latency bottleneck in flow matching VLA models, enabling real-time robotic control without sacrificing performance.
Full story at: arXiv Robotics ↗