CV — Ye Li

🎓 Education

📍 Shenzhen, ChinaSept 2023 – Present

Major: Computer Science and Technology
Labs: ITML Group & MMLab@SIGS
Group: leads the Efficient Deep Learning and Embodiment Group
Advisors: Prof. Shu-Tao Xia & Prof. Zhi Wang; co-advised by Prof. Yuan Meng
Research: Embodied intelligence (VLA, WAM, etc) & efficient deep learning for large models

📍 Tianjin, ChinaSept 2020 – Jun 2023

📍 Nanjing, ChinaSept 2016 – Jun 2020

💼 Experience

Huawei (华为)

Project Lead · Embodied-AI Data Preparation · industry research collaboration

📍 Shenzhen, ChinaAug 2025 – Aug 2026

Highlights

Led the full embodied-data-preparation project. Directed four workstreams — simulation platform, VLA algorithm development, VLA spatial-perception enhancement, and VLA inference acceleration.
Built the VLA inference-acceleration workstream. Optimized model execution to enable efficient collection of high-quality embodied data.

YXGN Robotics (远行光年)

Core Algorithm Member (intern) · embodied-AI startup incubated by our research group

📍 Shenzhen, ChinaOct 2025 – Mar 2026

Highlights

Built the lab's embodied-AI research platform that backed 10+ CCF-A submissions. Delivered a Franka + GELLO teleoperation stack — arm control, hand-eye calibration (eye-to-hand / eye-in-hand), and object detection & tracking.
Led a team to curate 10 TB+ of manipulation data. Developed the robot-arm data-collection system (recording, management, visualization) used for embodied-model training.

4Paradigm (第四范式)

Algorithm Intern

📍 Beijing, ChinaApr 2023 – Aug 2023

Highlights

Contributed to OpenRL, a unified open-source RL framework (800+ ★). Integrated RL algorithms and simulation environments, refactored the codebase, and improved stability & extensibility.
Developed multi-agent cooperative strategies for the Jidi (及第) football competition. Built and validated policies on the team's TMARL framework, covering policy training and environment adaptation.

📄 Selected Publications * = first / co-first author

First / co-first author

Preprint
2026

Y. Li*, H. Liu, K. Ji, Y. Meng, J. Fan, Y. Wang, S. Qin, C. Wu, S.-T. Xia, Z. Wang

⚡ up to 2.55× (GR00T) · 3.77× (CogACT) · real-world (Franka) 13.8→26.3 Hz

ICLR
2026

Y. Li*, Y. Meng, Z. Sun, K. Ji, C. Tang, J. Fan, X. Ma, S.-T. Xia, Z. Wang, W. Zhu

⚡ 2.5x lossless (Franka) 1.5× lossless (LIBERO) · 2.4× (SimplerEnv)

Preprint
2025