# TouchBridge: Tactile-Conditioned Human-to-Robot Transfer for Dexterous Manipulation

**Authors:** Haozhe Zhang, Gu Zhang, Zirui Zou, Yuzheng Chen, Jinlong Wu, Zixuan Hu, Jianke Zhang, Yucheng Hu, Zhecheng Yuan, Huazhe Xu  
**Lab:** Tsinghua University · **Published:** 29 Oct 2026  
**Links:** [arXiv](https://arxiv.org/abs/2610.40349) · [PDF](https://arxiv.org/pdf/2610.40349) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2610.40349.html)

## Key numbers

- **72.5%**: policies fine-tuned solely on TouchBridge-converted human demonstrations achieve a 72.5% average success rate
- **77.5%**: compared with 77.5% using an equal number of teleoperated robot demonstrations
- **7.0×**: while human demonstration collection is 7.0x faster

## Abstract

Human demonstrations offer an efficient and scalable source of data for dexterous robot manipulation, but differences in sensing and hand morphology complicate the transfer of contact-rich interactions. We present TouchBridge, a hardware--software co-designed framework for human-to-robot data transformation. A wearable human demonstration interface and a corresponding robot interface establish cross-embodiment correspondence in visual and fingertip tactile sensing. We introduce Tac-Retarget, a tactile-conditioned retargeting method that uses measured contact to align fingertip contact patches and preserve their spatial relations within and across hands. Unlike object-aware or simulation-based transfer pipelines, Tac-Retarget requires no object mesh, object-pose estimate, or physics simulation for retargeting, enabling direct and scalable conversion of human demonstrations into robot-domain training data. Combined with visual synthesis and camera-anchored action conversion, TouchBridge produces robot-domain demonstrations for fine-tuning a pretrained vision-language-action (VLA) model. Across four real-world contact-rich tasks, policies fine-tuned solely on TouchBridge-converted human demonstrations achieve a 72.5% average success rate, compared with 77.5% using an equal number of teleoperated robot demonstrations, while human demonstration collection is 7.0x faster. Under the same policy and training setup, Tac-Retarget outperforms DexPilot and UniDex by 25.0 and 42.5 percentage points, respectively.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.