Siqi Song
Logo Undergraduate Student, Tsinghua University

I am currently a senior undergraduate at Department of Electronic Engineering, Tsinghua University. Previously, I had the privilege to work as an undergraduate research intern under the supervision of Prof. Yue Wang.

I am currently seeking for an Ph.D./R.A. position starting from Fall 2026. Feel free to drop me an email if you are interested in my research or have any questions.

My interests focus on Robotic Intelligence and Reinforcement Learning, in particular on the following aspects:

  • Reinforcement Learning: How Reinforcement Learning scales and how can this boost robotic learning?
  • Advancing Robot-Environment-Human Interaction: With the intricate dynamics of reality, how can robots seamlessly engage in complex yet meaningful interactions with the environment and humans?
  • Learning the Embodiment and Understanding the Environment: How do robotic embodiments grasp their own physical capabilities and properties, as well as the characteristics of the surrounding environment? Furthermore, how does this knowledge shape their decision-making processes and execution strategies?
  • Multi-agent Collaboration: How can multiple embodiments work together to tackle complex problems?

Publications (view all )
Universal Humanoid Robot Pose Learning from Internet Human Videos
Universal Humanoid Robot Pose Learning from Internet Human Videos

Jiageng Mao*, Siheng Zhao*, Siqi Song*, Chuye Hong, Tianheng Shi, Junjie Ye, Mingtong Zhang, Haoran Geng, Jitendra Malik, Vitor Guizilini, Yue Wang (* equal contribution)

International Conference on Humanoid Robots (Humanoids) 2025

We propose the first scalable approach for learning generalizable and universal pose control for humanoid robots using Internet videos, which includes Humanoid-X, a massive dataset of 20M+ robot poses with text descriptions, and UH-1, a language-driven motion generation model.

Universal Humanoid Robot Pose Learning from Internet Human Videos

Jiageng Mao*, Siheng Zhao*, Siqi Song*, Chuye Hong, Tianheng Shi, Junjie Ye, Mingtong Zhang, Haoran Geng, Jitendra Malik, Vitor Guizilini, Yue Wang (* equal contribution)

International Conference on Humanoid Robots (Humanoids) 2025

We propose the first scalable approach for learning generalizable and universal pose control for humanoid robots using Internet videos, which includes Humanoid-X, a massive dataset of 20M+ robot poses with text descriptions, and UH-1, a language-driven motion generation model.

All publications
Education
  • Tsinghua University
    Tsinghua University
    B.S. in Electonic Engineering
    Sep. 2021 - Present
Honors & Awards
  • Chinese National Scholarship
    2024
  • Scholarship for Comprehensive Excellence, Tsinghua University
    2023
  • Scholarship for Comprehensive Excellence, Tsinghua University
    2022