Adaptive action chunking improves robot control by varying horizon based on prediction reliability.
GeoAAC uses the geometry of flow matching denoising trajectories to adjust the action chunk size per step, boosting real-world success rates from 53.3% to 74.4%.
Industry
Xin Chen · Sen Chen · Yujuan Ding · Jian Liu · Guoqing Wang · Wei Ye · +2 more
Tongji University · The Hong Kong Polytechnic University · University of Electronic Science and Technology of China
Research Digest··2 min read
Thread:Live Software Adaptation
The authors propose GeoAAC, a method for Vision-Language-Action (VLA) policies that adaptively sets the action horizon—the number of future actions predicted at once—based on the geometric properties of the denoising process.
Why this paper
From Tongji University and 2 others · Part of Live Software Adaptation, now 4 papers
In one line
GeoAAC adapts action chunking horizon from denoising trajectory geometry, improving VLA policy success rates over fixed horizons.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§