aiqbench analysis
This research demonstrates that integrating tactile sensing into Vision-Language-Action models can be achieved through a single-stage training approach, eliminating the need for separate policy pretraining or alignment stages. By leveraging frozen pretrained encoders and action supervision, SimpleTouch significantly improves success rates on contact-rich manipulation tasks compared to existing methods. This finding suggests a more efficient pathway for deploying robust robotic manipulation systems without incurring substantial additional data or computational costs.
SimpleTouch: Can Vision-Language-Action Models Master Contact-Rich Manipulation Without Tactile Policy Pretraining?
arXiv:2610.02784v1 Announce Type: new Abstract: Tactile sensing provides essential contact information for robotic manipulation, yet incorporating it into pretrained vision-language-action (VLA) models remains challenging. A common concern is that simply introducing touch during task-specific fine-tuning may fail to bridge the cross-modal gap, yielding limited gains or even reduced success. Consequently, existing methods often rely on large-scale tactile policy pretraining or separate visuotactile alignment, adding data requirements and training stages. We introduce SimpleTouch, a simple VLA extension that augments $\pi_{0.5}$ with a tactile expert, to test whether these additional stages are necessary. Leveraging all tokens from a frozen pretrained tactile encoder, the expert learns from action supervision and multi-horizon prediction of future tactile latents. This single-stage training uses only task demonstrations, without additional tactile policy pretraining or separate alignment. With 50 demonstrations per task, SimpleTouch achieves the highest success rate among evaluated methods on all six UniVTAC tasks. Its average success rate reaches 77.5%, compared with 45.2% for FTP-$\pi_{0.5}$ and 66.7% for FTP-1, corresponding to gains of 32.3 and 10.8 percentage points, respectively. Across four real-world tasks, it averages 71.3%, exceeding FTP-1 by 8.8 percentage points. These results demonstrate that, given pretrained VLA and tactile representations, additional tactile policy pretraining is not a prerequisite for strong performance on these tasks, offering a simpler route to contact-rich manipulation. Project page: https://simpletouch-robot.github.io/
Title:SimpleTouch: Can Vision-Language-Action Models Master Contact-Rich Manipulation Without Tactile Policy Pretraining?
Abstract:Tactile sensing provides essential contact information for robotic manipulation, yet incorporating it into pretrained vision-language-action (VLA) models remains challenging. A common concern is that simply introducing touch during task-specific fine-tuning may fail to bridge the cross-modal gap, yielding limited gains or even reduced success. Consequently, existing methods often rely on large-scale tactile policy pretraining or separate visuotactile alignment, adding data requirements and training stages. We introduce SimpleTouch, a simple VLA extension that augments $\pi_{0.5}$ with a tactile expert, to test whether these additional stages are necessary. Leveraging all tokens from a frozen pretrained tactile encoder, the expert learns from action supervision and multi-horizon prediction of future tactile latents. This single-stage training uses only task demonstrations, without additional tactile policy pretraining or separate alignment. With 50 demonstrations per task, SimpleTouch achieves the highest success rate among evaluated methods on all six UniVTAC tasks. Its average success rate reaches 77.5%, compared with 45.2% for FTP-$\pi_{0.5}$ and 66.7% for FTP-1, corresponding to gains of 32.3 and 10.8 percentage points, respectively. Across four real-world tasks, it averages 71.3%, exceeding FTP-1 by 8.8 percentage points. These results demonstrate that, given pretrained VLA and tactile representations, additional tactile policy pretraining is not a prerequisite for strong performance on these tasks, offering a simpler route to contact-rich manipulation. Project page: this https URL
Robotics (cs.RO)
arXiv:2610.02784 [cs.RO]
(or arXiv:2610.02784v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.02784 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 320 words
Related coverage
Don't Drop the BATON: Long-Horizon Robot Manipulation via Agentic Subtask Exploration and Transition-aware Memory
arXiv · Aug 18, 2026
Bridging the Morphology Gap: Adapting VLA Models to Dexterous Manipulation via Intent-Conditioned Fine-Tuning
arXiv · Jun 11, 2026
ForeTac-VLA: A Forecasting-Based Tactile-Vision-Language-Action Model for Contact-Rich Robotic Manipulation
arXiv · Sep 21, 2026
VT-Bridge: Bridging Pretrained Foundation VLAs to VTLAs via Lightweight Residual Adaptation
arXiv · Sep 22, 2026