Kaifeng Zhang

I am a Research Scientist at Sharpa Robotics. I have previously worked for NetEase AI Lab, Tencent Robotics X, Huawei (Autonomous Driving) and Shanghai Qi Zhi Institute.

My work lies in machine learning, computer vision and robotics. In particular, my research focuses on solving (1) high cost for training data in robotic domain, (2) high cost for policy learning in the real-world, and (3) high cost for model evaluation in the real-world.

Email / Twitter

Selected Publications

Representative papers are highlighted.

Acceptance builds resume, while rejection makes man.

Embodied Playground: A Long-Horizon Agentic Robotics Benchmark for Physical, Strategic, and Spatial Reasoning
Feng Chen*, Pei Zhou*, Lixuan He*, Yunchao Yao*, Li Sun*, Shenghua Gao, Mingyu Ding, Kaifeng Zhang, Kaichen Zhou, Yi Ma
arXiv

Unified Visual-Tactile-Action Modeling from Human Demonstrations for Dexterous Manipulation
Wenqiao Li, Qianyou Zhao, Jiawen Hao, Xuezhou Zhu, Tengyu Liu, Kaifeng Zhang, Chuan Wen^†, Siyuan Huang^†
arXiv

TacBPM: A Tactile-conditioned Behavior Prior Model for Dexterous Reorientation
Jie Yin, Wanli Xing, Zeyuan Zhao, Xuezhou Zhu, Zhijie Deng, Kaifeng Zhang^†
arXiv

Grounding Generated Video Plans in Simulation Towards Versatile Dexterous Controllers
Tianyue Wu*, Boyuan An*, Shuqi Zhao, Heyu Guo, Wanli Xing, Yi Ma, Kaifeng Zhang, Ruihai Wu^†, Masayoshi Tomizuka^†
arXiv

R2S-Eval: Robot Evaluation with Real-to-Sim Calibration via Vision-Language Models
Yidi Wang*, Feixiang Ruan*, Ruoqu Chen, Jie Yin, Yang Yu, Mengdi Xu, Kaifeng Zhang^†
arXiv

Learning In-Hand Object Reaching to General 6D Poses
Junxiao Lin, Tianyue Wu, Jie Yin, Jia Pan, Kaifeng Zhang, Weiming Zhi
arXiv

FTP-1: A Generalist Foundation Tactile Policy Across Tactile Sensors for Contact-Rich Manipulation
Chengbo Yuan*, Zicheng Zhang*, Mingjie Zhou*, Wendi Chen, Yi Wang, Zhuoyang Liu, Dantong Niu, Shuo Wang, Hui Zhang, Wenkang Zhang, Yingdong Hu, Yuanqing Gong, Wanli Xing, Chuan Wen, Cewu Lu, Kaifeng Zhang^†, Yang Gao^†
CoRL, 2026
arXiv

A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms
UniLab Team
CoRL, 2026
arXiv

Dex-x: Learning Visual-Tactile Dexterous Manipulation From Human Demonstration with Simulated Interaction
Ruoqu Chen, Feixiang Ruan, Liu Cao, Zihao Wang, Botian Xu, Shiqin Tong, Jiajun Liu, Wanli Xing, Kaifeng Zhang, Mengdi Xu
CoRL, 2026
arXiv

Towards Human-Like Manipulation through RL-Augmented Teleoperation and Mixture-of-Dexterous-Experts VLA
Tutian Tang*, Xingyu Ji*, Wanli Xing*, Ce Hao, Wenqiang Xu, Lin Shao, Jiangmiao Pang, Cewu Lu, Qiaojun Yu^†, Kaifeng Zhang^†
IROS, 2026
arXiv

DexEMG: Towards Dexterous Teleoperation System via EMG2Pose Generalization
Qianyou Zhao, Wenqiao Li, Chiyu Wang, Kaifeng Zhang^†
IROS, 2026
arXiv

Tacmap: Bridging the Tactile Sim-to-Real Gap via Geometry-Consistent Penetration Depth Map
Lei Su*, Zhijie Peng*, Renyuan Ren, Shengping Mao, Juan Du, Kaifeng Zhang, Xuezhou Zhu
arXiv

Spatially Anchored Tactile Understanding for Robust Dexterous Manipulation
Jialei Huang*, Yang Ye*, Yuanqing Gong, Xuezhou Zhu, Yang Gao^†, Kaifeng Zhang^†
ICRA, 2026 Oral
arXiv

ViTacFormer: Learning Cross-Modal Representation for Visuo-Tactile Dexterous Manipulation
Liang Heng*, Haoran Geng*, Kaifeng Zhang, Pieter Abbeel, Jitendra Malik
RSS, 2026
arXiv

Learning Manipulation Skills through Robot Chain-of-Thought with Sparse Failure Guidance
Kaifeng Zhang, Zhao-Heng Yin, Weirui Ye, Yang Gao
IROS, 2025
arXiv / code

Summary:
(1) We propose that vision-language models can provide a superior source of rewards for agents..
(2) Our algorithm achieves a 5.4x higher average success rate compared to the best baseline, RoboCLIP.

Auto-Encoding Adversarial Imitation Learning
Kaifeng Zhang, Rui Zhao, Ziming Zhang, Yang Gao
AAMAS, 2024
arXiv / code

Summary:
(1) To the best of our knowledge, we propose the first formulation of auto-encoder based reward signal for adversarial imitation learning.
(2) Our auto-encoding AIL provides a denser reward signal to the agent policy.

心外无物莫向外求