TempoBridge enables robots to adjust action speed from language cues
TempoBridge: Language-Guided Tempo Control for Vision-Language-Action Policies
Robotics
Summary
Robots that understand what to do often can’t easily change how fast they do it based on instructions. The authors present TempoBridge, which lets robots use language hints about speed to control their movements during tasks without new training. It works by reading tempo cues in the instruction, matching them to task progress, and adjusting movement speed accordingly. Their experiments show TempoBridge helps robots better follow speed instructions while still completing tasks successfully, including in real-world robot tests.
What this means in practice
- •For robotics engineers: Incorporate TempoBridge to enable robots to adjust motion speed based on verbal tempo instructions during complex tasks.
- •For automation system designers: Design automated workflows that flexibly control operation speed from natural language without needing specialized training data.
Authors
Yeonseo Lee, Hyosup Shin, Guebin Hwang, Sungho Jo
Abstract
Vision-Language-Action (VLA) models are effective at understanding what task to perform, but provide limited control over how it should be executed, such as moving quickly or slowly. We introduce TempoBridge, a lightweight framework that uses frozen VLA representations to modulate actions according to tempo cues in the instruction at each task phase, without additional tempo-conditioned robot demonstrations or tempo-specific base-policy fine-tuning. TempoBridge extracts tempo cues from contextual VLM representations, aligns them with task progress through a causal phase router, and modulates nominal motion commands during execution. Across LIBERO tasks, TempoBridge improves Tempo Success Rate from 52.6% to 89.7% under canonical tempo instructions while retaining high task success. It also preserves near-baseline performance when no tempo cue is present and generalizes to unseen tempo expressions without additional training. Experiments on a physical robot further demonstrate language-conditioned tempo modulation in real-world manipulation.