
XG
Xudi Ge, Yepeng Liu, Haifeng Liu, Enze Shi, Liangqi Zhang, Yongchao Xu
· 1 min read
ResearcharXiv cs.AI
TipDex-VTLA: Learning When to Refine Actions with Touch for Dexterous Manipulation
arXiv:2611.25683v1 Announce Type: cross
Abstract: Changes in illumination can induce action drift in dexterous manipulation even when the physical state remains unchanged. Although existing vision--tactile--language--action (VTLA) policies incorporate tactile feedback, when and how touch refines actions under changing visual conditions remains challenging. We introduce \textbf{TipDex-VTLA}, a VTLA policy that combines estimated visual action reliability and tactile forecast verification to regulate execution-time refinement. Specifically, Visual Action Sensitivity Estimation (VASE) learns visual reliability from a frozen reference policy's action sensitivity to illumination perturbations. Tactile Anticipation and Correction during Execution (TACE) compares predicted and measured touch at matched times and combines the resulting discrepancy with visual reliability to gate bounded feature corrections for unexecuted actions. This effectively helps to refine actions under challenging unseen visual conditions. We evaluate TipDex-VTLA on four real-world dexterous manipulation tasks and the UniVTAC simulation benchmark. TipDex-VTLA trained on data collected under normal visual conditions, achieves average full-task success rates of 87.5% (resp. 73.8%) under normal (resp. challenging unseen) lighting in real-world tasks, significantly outperforming some state-of-the-art methods. We will release the collected data and source code.
Original source
This story was published by arXiv cs.AI and written by Xudi Ge, Yepeng Liu, Haifeng Liu, Enze Shi, Liangqi Zhang, Yongchao Xu. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


