GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • Chen Sirui Eric
  • Publications
  • Talks
  • Teaching
  • CV

This view

  • Original page
  • New address

GeezerVison is showing ericcsr.github.io — About me

Hi, I am Sirui (Eric) Chen (陈思睿), a final-year CS PhD student at Stanford advised by Professor Karen Liu. I am interested in designing robot control and planning algorithms for contact-rich tasks such as dexterous manipulation and humanoid whole-body loco-manipulation. I also work on solving complex control problems on real robots by combining human demonstrations from different sources with analytical modeling.

I interned at Honda Research Institute in summer 2024, advised by Rana Soltani Zarrin; at NVIDIA GEAR Lab in summer and fall 2025, advised by Yuke Zhu and Zhengyi Luo; and at Amazon FAR in 2026, advised by Karen Liu and Guanya Shi.

In my free time, I enjoy photography and hiking; here are some photos I have taken. Feel free to reach out about full-time opportunities or to discuss research ideas.

Selected research

SceneBot: Contact-Prompted General Humanoid Whole Body Tracking with Scene-Interaction

SceneBot: Contact-Prompted General Humanoid Whole Body Tracking with Scene-Interaction

Arxiv, 2026

A unified contact-conditioned humanoid tracker for free-space motion, terrain, and object interaction

Paper · Website

CHIP: Adaptive Compliance for Humanoid Control through Hindsight Perturbation

CHIP: Adaptive Compliance for Humanoid Control through Hindsight Perturbation

Conference on Robot Learning, 2026

A plug-and-play module for controllable humanoid end-effector compliance

Paper · Website

SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control

SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control

Science Robotics, 2026

Scaling up humanoid robot control via motion tracking

Paper · Website

All publications

You are visitor 01402503 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.