Ruohan Zhang (@RuohanZhang76) / X

Ruohan Zhang

191 posts

Ruohan Zhang

@RuohanZhang76

Incoming Assistant Professor @NorthwesternCS, Postdoc @StanfordSVL; robot, brain, art; soccer, cooking, dance

Stanford University

ruohanzhang.com

Joined September 2021

Pinned
Ruohan Zhang
@RuohanZhang76
21h
The 2nd BEHAVIOR Challenge is here! As embodied AI models advance at a rapid pace, it's more important than ever to understand where the field stands on generalizable solutions. With high-quality teleoperation data and strong pre-trained baselines including π₀ and GR00T, we
Fei-Fei Li
@drfeifei
Jul 13
1/N Long horizon, complex tasks that truly matter in everyday life are not solved problems by today’s robotics, requiring planning, object detection, object manipulation, and failure recovery. That's why Stanford's BEHAVIOR Challenge is back for year 2! Last year, the winning
00:00
2.6K
Ruohan Zhang
@RuohanZhang76
Nov 9, 2025
I will join Northwestern University Computer Science as an Assistant Professor in Fall 2026! I am actively recruiting PhD students and seeking collaborations in robotics, human-robot interaction, brain-computer interfaces, cognitive science, societal impact of AI & automation,
612K
Ruohan Zhang
@RuohanZhang76
Nov 3, 2023
Introducing our new work @corl_conf 2023, a novel brain-robot interface system: NOIR (Neural Signal Operated Intelligent Robots). Website: noir-corl.github.io Paper: arxiv.org/abs/2311.01454 🧠🤖
00:00
288K
Ruohan Zhang
@RuohanZhang76
Sep 2, 2024
I believe in big data for robotics. But in this work with @wenlong_huang , he taught me this important lesson again: the right representation + foundation models + constrained optimization can enable robots to perform very challenging tasks without task-specific training data.
Wenlong Huang
@wenlong_huang
Aug 29, 2024
What structural task representation enables multi-stage, in-the-wild, bimanual, reactive manipulation? Introducing ReKep: LVM to label keypoints & VLM to write keypoint-based constraints, solve w/ optimization for diverse tasks, w/o task-specific training or env models. 🧵👇
00:00
21K
Ruohan Zhang
@RuohanZhang76
Sep 9, 2025
Thanks to everyone’s interest in BEHAVIOR so far! We have received several questions, and I am trying to answer some of them here: 1. 📜How are tasks defined in BEHAVIOR? BEHAVIOR tasks are written in BDDL (BEHAVIOR Domain Definition Language). Unlike geometric, image/video, or
Fei-Fei Li
@drfeifei
Sep 2, 2025
(1/N) How close are we to enabling robots to solve the long-horizon, complex tasks that matter in everyday life? 🚨 We are thrilled to invite you to join the 1st BEHAVIOR Challenge @NeurIPS 2025, submission deadline: 11/15. 🏆 Prizes: 🥇 $1,000 🥈 $500 🥉 $300
00:00
17K
Ruohan Zhang
@RuohanZhang76
Mar 22, 2024
BEHAVIOR is my first project since I join Stanford and the largest academic research project I co-lead. I am grateful to be among 50+ students, engineers, and faculty who contributed. Please stay tuned for the next step of BEHAVIOR!
Fei-Fei Li
@drfeifei
Mar 22, 2024
One year ago, we first introduced BEHAVIOR-1K, which we hope will be an important step towards human-centered robotics. After our year-long beta, we’re thrilled to announce its full release, which our team just presented at NVIDIA #GTC2024. 1/n
00:00
8.2K
Ruohan Zhang
@RuohanZhang76
Oct 24, 2025
One advantage of collecting human demonstration data in simulation is that, we can effectively use motion planning to generate 1000x more data. MoMaGen is a follow-up of the X-Gen series, enabling data augmentation in long-horizon mobile manipulation tasks. See Eric's post
Chengshu Li
@ChengshuEricLi
Oct 23, 2025
We are excited to release MoMaGen, a data generation method for multi-step bimanual mobile manipulation. MoMaGen turns 1 human-teleoped robot trajectory into 1000s of generated trajectories automatically.🚀 Website: momagen.github.io arXiv: arxiv.org/abs/2510.18316
00:00
7.2K
Ruohan Zhang
@RuohanZhang76
Mar 4, 2025
Please stay tuned for our upcoming release for this fun project!
Yunfan Jiang
@YunfanJiang
Mar 4, 2025
🚀Two weeks ago, we hosted a welcome party for the newest member of our Stanford Vision and Learning Lab—a new robot! 🤖✨Watch as @drfeifei interacts with it in this fun video. Exciting release coming soon. Stay tuned! 👀🎉
00:00
3.3K
Ruohan Zhang
@RuohanZhang76
Nov 3, 2023
Replying to @RuohanZhang76
Envisioned by many, brain-robot interface (BRI) stands out as a thrilling but challenging research topic. It is an exciting time for BRI research. Brain signal decoding and robot intelligence are improved a lot by modern machine learning algorithms.
3.8K
Ruohan Zhang
@RuohanZhang76
Nov 10, 2024
We have seen many exciting robotics works leveraging LLMs. How good are they at solving robotic tasks? What's needed is a powerful tool that allows us to perform controlled experiments and in-depth analyses. Embodied Agent Interface is exactly what you need to achieve this, 👇
Manling Li
@ManlingLi_
Nov 6, 2024
[NeurIPS D&B Oral] Embodied Agent Interface: Benchmarking LLMs for Embodied Agents A single line of code to evaluate your model! 🌟Standardize Goal Specifications: LTL 🌟Standardize Modules and Interfaces: 4 modules, 438 tasks, 1475 goals 🌟Standardize Fine-grained Metrics: 18
00:00
3.5K
Ruohan Zhang
@RuohanZhang76
Nov 3, 2023
Replying to @RuohanZhang76
NOIR is a general-purpose, intelligent BRI system that enables humans to command robots to perform 20 challenging everyday activities using their brain signals, such as cooking, cleaning, playing games with friends, and petting a (robot) dog.
00:00
5K
Ruohan Zhang
@RuohanZhang76
Mar 13, 2025
Since we released BEHAVIOR, some critical questions remain: 1) To accomplish 1000 household tasks, what key capabilities should a robot have? 2) What hardware designs are necessary? 3) How do we efficiently collect demo data for such hardware? 4) How do we learn from such data?
Yunfan Jiang
@YunfanJiang
Mar 10, 2025
🤖 Ever wondered what robots need to truly help humans around the house? 🏡 Introducing 𝗕𝗘𝗛𝗔𝗩𝗜𝗢𝗥 𝗥𝗼𝗯𝗼𝘁 𝗦𝘂𝗶𝘁𝗲 (𝗕𝗥𝗦)—a comprehensive framework for mastering mobile whole-body manipulation across diverse household tasks! 🧹🫧 From taking out the trash to
00:00
4.3K
Ruohan Zhang
@RuohanZhang76
Apr 3, 2024
It’s a great honor to be an co-instructor for 231n, the course that I have watched so many times on YouTube. Looking forward to working with the team and giving lectures!
Fei-Fei Li
@drfeifei
Apr 3, 2024
It’s that time of the year - first lecture of @cs231n !! It’s the 9th year since @karpathy and I started this journey in 2015, what an incredible decade of AI and computer vision! Am so excited to this new crop of students in CS231n! (Co-instructing with @eadeli this year 😍🤩)
3.4K
Ruohan Zhang
@RuohanZhang76
Nov 3, 2023
Replying to @RuohanZhang76
With 10-minute calibration for each session, 3 human participants successfully accomplished 20 long-horizon tasks (4-15 subtasks). On average, each task requires 1.8 attempts to succeed with an average task completion time of 20 minutes.
1.7K