KAIST's VSpark Speeds Up Autoregressive VLA Robot Models 2.1x with Speculative Decoding

kaist-ai · hf · 2026-11-20

KAIST researchers present VSpark, bringing speculative decoding to autoregressive vision-language-action (VLA) models with lossless 1.28x–2.10x speedups on robot manipulation benchmarks.

A practical step toward lower-latency robot policy inference.

Original post →

More from Embodied

Embodied channel →