One Euler step suffices: VLAs get 1.6-4.8x faster with no retraining
kaixhin · x · 2026-11-10
Work to be presented at CoRL 2026 shows VLA models, which typically run 4-10 Euler steps per action, can largely match multi-step performance with just 1 step.
The finding was validated across 6 VLAs, 4 simulation benchmarks, and 4.8k real-robot rollouts, delivering 1.6-4.8x faster inference without any retraining or distillation.
More from Embodied
- HermesEnabler2 confirmed working on Apple Watch Series 10, unlocking Hermès watch faces — gregbarbosa · 2026-11-10
- Fable 5.2 hits 64.9% on RoboDojo, beating purpose-built robot AI systems — CyberRobooo · 2026-11-10
- Caltech robot builds a structure in orbit autonomously, a first for space construction — GregCook2011 · 2026-11-10
- Robot paints original watercolors and ships them with telemetry — a one-man robot art club — repligate · 2026-11-10
- Day 5 of a DIY robot arm: CAD render of the differential wrist in progress — _Stocko_ · 2026-11-10
- SuperMimic's 30× gradient SNR edge over score-function cuts imagined steps 64× vs PPO — daibond_alpha · 2026-11-10