FASTER: Rethinking Real-Time Flow VLAs
2026-03-19 • Robotics
RoboticsComputer Vision and Pattern Recognition
AI summaryⓘ
The authors focus on improving how fast Vision-Language-Action (VLA) models react to changes in their environment, which is important for real-time robot actions. They found that current methods waste time by waiting too long before starting movement, causing delays. To fix this, they developed FASTER, a new approach that speeds up immediate reactions by prioritizing near-term actions while still planning ahead. Tests on real robots, like in table tennis, showed this method makes responses much quicker without losing accuracy.
Vision-Language-Action modelsreaction timeaction chunkingflow-based samplingdenoisingtrajectory planningreal-time roboticslatencyHorizon-Aware Schedulerobotics control
Authors
Yuxiang Lu, Zhe Liu, Xianzhe Fan, Zhenya Yang, Jinghua Hou, Junyi Li, Kaixin Ding, Hengshuang Zhao
Abstract
Real-time execution is crucial for deploying Vision-Language-Action (VLA) models in the physical world. Existing asynchronous inference methods primarily optimize trajectory smoothness, but neglect the critical latency in reacting to environmental changes. By rethinking the notion of reaction in action chunking policies, this paper presents a systematic analysis of the factors governing reaction time. We show that reaction time follows a uniform distribution determined jointly by the Time to First Action (TTFA) and the execution horizon. Moreover, we reveal that the standard practice of applying a constant schedule in flow-based VLAs can be inefficient and forces the system to complete all sampling steps before any movement can start, forming the bottleneck in reaction latency. To overcome this issue, we propose Fast Action Sampling for ImmediaTE Reaction (FASTER). By introducing a Horizon-Aware Schedule, FASTER adaptively prioritizes near-term actions during flow sampling, compressing the denoising of the immediate reaction by tenfold (e.g., in $π_{0.5}$ and X-VLA) into a single step, while preserving the quality of long-horizon trajectory. Coupled with a streaming client-server pipeline, FASTER substantially reduces the effective reaction latency on real robots, especially when deployed on consumer-grade GPUs. Real-world experiments, including a highly dynamic table tennis task, prove that FASTER unlocks unprecedented real-time responsiveness for generalist policies, enabling rapid generation of accurate and smooth trajectories.