RESEARCH / PUBLICATIONS

Selected research.

A collection of published work in robot learning, embodied AI, and vision-language-action models.

PUBLICATION RECORD3 SELECTED PAPERS / 2025—2026

ZETA experimental setup and cross-embodiment generalization diagram

CoRL 2026SPOTLIGHT2026

ZETA: A Controlled Study of Zero-Shot Cross-Embodiment VLA Transfer for Tabletop Manipulation

Mi Yan*, Wenhao Zhang*, Zhiqi Zhang*, Yu Peng*, Tangxinyu Wang*, Lingfei Zhai, Jiayi Su, Shengliang Deng, Lin Peng, Yaowei Liu, Yuxing Chen, Zhiyuan Wei, Jilong Wang, Jiayi Chen, Jiangran Lyu, Zhizheng Zhang†, He Wang†

A controlled study of cross-embodiment VLA transfer, separating strict zero-shot generalization to unseen robot embodiments from pretrain-exposed transfer, and analyzing the effects of state-action representations, embodiment diversity, auxiliary co-training, and target-embodiment exposure.

Paper Project

StereoVLA robot camera configurations and depth perception challenges

RSS 20262026

StereoVLA: Enhancing Vision-Language-Action Models with Stereo Vision

Shengliang Deng*, Mi Yan*, Yixin Zheng*, Jiayi Su, Wenhao Zhang, Xiaoguang Zhao, Heming Cui, Zhizheng Zhang†, He Wang†

A stereo-vision VLA that jointly models semantic and geometric cues for fine-grained spatial perception, improving manipulation accuracy and robustness across large camera-pose variations.

Paper Project Code

GraspVLA pretraining and generalization research teaser

CoRL 20252025

GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Data

Shengliang Deng*, Mi Yan*, Songlin Wei, Haixin Ma, Yuxin Yang, Jiayi Chen, Zhiqi Zhang, Taoyu Yang, Xuheng Zhang, Wenhao Zhang, Heming Cui, Zhizheng Zhang†, He Wang†

A grasping foundation model pretrained on billion-scale synthetic action data, enabling direct sim-to-real transfer, open-vocabulary generalization, and few-shot adaptation to downstream preferences.

Paper Project Code

* Equal contribution † Corresponding author