Mengdi Xu Profile
Mengdi Xu

@mengdixu_

Followers
2K
Following
914
Media
6
Statuses
80

Assistant Prof. @Tsinghua_Uni. Postdoc @StanfordSVL. Ph.D. @CarnegieMellon. Prev. @GoogleDeepMind. Learning and Robotics.

Joined November 2017
Don't wanna be here? Send us removal request.
@mengdixu_
Mengdi Xu
28 days
How can we scale robot data to cover diverse household scenarios? ๐Ÿก๐ŸณOne promising direction is generating large-scale bimanual mobile manipulation data in simulation! Excited to introduce MoMaGen, which is a scalable pipeline that automatically generates diverse long-horizon
@ChengshuEricLi
Chengshu Li
28 days
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.๐Ÿš€ Website: https://t.co/DYKvqY4bII arXiv: https://t.co/lDffi0FXHl
2
8
81
@drfeifei
Fei-Fei Li
10 days
AIโ€™s next frontier is Spatial Intelligence, a technology that will turn seeing into reasoning, perception into action, and imagination into creation. But what is it? Why does it matter? How do we build it? And how can we use it? Today, I want to share with you my thoughts on
168
626
3K
@RuohanZhang76
Ruohan Zhang
11 days
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,
67
207
1K
@RuohanZhang76
Ruohan Zhang
2 months
Thanks to everyoneโ€™s interest in BEHAVIOR so far! We have received several questions, and I am trying to answer some of them here: 1. ๐Ÿ“œHow are tasks defined in BEHAVIOR? BEHAVIOR tasks are written in BDDL (BEHAVIOR Domain Definition Language). Unlike geometric, image/video, or
behavior-robot-suite.github.io
Streamlining Real-World Whole-Body Manipulation for Everyday Household Activities
@drfeifei
Fei-Fei Li
3 months
(1/N) How close are we to enabling robots to solve the long-horizon, complex tasks that matter in everyday life? ๐Ÿšจ We are thrilled to invite you to join the 1st BEHAVIOR Challenge @NeurIPS 2025, submission deadline: 11/15. ๐Ÿ† Prizes: ๐Ÿฅ‡ $1,000 ๐Ÿฅˆ $500 ๐Ÿฅ‰ $300
3
14
103
@mengdixu_
Mengdi Xu
3 months
Very excited to announce the 1st BEHAVIOR Challenge! ๐Ÿš€๐Ÿค– With so many recent advances, how well do todayโ€™s general-purpose models handle household tasks that are both practical and technically challenging? We invite you to submit your work and help benchmark the field. Can your
@drfeifei
Fei-Fei Li
3 months
(1/N) How close are we to enabling robots to solve the long-horizon, complex tasks that matter in everyday life? ๐Ÿšจ We are thrilled to invite you to join the 1st BEHAVIOR Challenge @NeurIPS 2025, submission deadline: 11/15. ๐Ÿ† Prizes: ๐Ÿฅ‡ $1,000 ๐Ÿฅˆ $500 ๐Ÿฅ‰ $300
1
7
63
@mengdixu_
Mengdi Xu
5 months
Excited to share that ROSETTA won the Best Paper Award at the CRLH workshop @RSS! ๐ŸŽ‰ Huge kudos to the team, and many thanks to the organizers for a fantastic workshop! Iโ€™ll also be at the HitLRL workshop today. Happy to chat about building robots that better understand, assist,
0
0
11
@sanjana__z
Sanjana Srivastava
5 months
ROSETTA won best paper at CRLH @ RSS and will be at HitLRL tomorrow! Thanks to @James_KKW, Jerry Chan, Roger Dai, @ManlingLi_, @mengdixu_, @RuohanZhang76, @jiajunwu_cs, and @drfeifei! Website: https://t.co/Sel0u8XoUB Code: https://t.co/A45CnR1HAK Paper:
0
5
21
@mengdixu_
Mengdi Xu
5 months
Iโ€™ve always been thinking about how to make robots naturally co-exist with humans. The first step is having robots understand our unconstrained, dynamic preferences and follow them. ๐Ÿค We proposed ROSETTA, which translates free-form language instructions into reward functions to
@sanjana__z
Sanjana Srivastava
5 months
๐Ÿค– Household robots are becoming physically viable. But interacting with people in the home requires handling unseen, unconstrained, dynamic preferences, not just a complex physical domain. We introduce ROSETTA: a method to generate reward for such preferences cheaply. ๐Ÿงตโฌ‡๏ธ
4
5
52
@wenlong_huang
Wenlong Huang
6 months
How to scale visual affordance learning that is fine-grained, task-conditioned, works in-the-wild, in dynamic envs? Introducing Unsupervised Affordance Distillation (UAD): distills affordances from off-the-shelf foundation models, *all without manual labels*. Very excited this
9
112
437
@ZeYanjie
Yanjie Ze
7 months
๐Ÿค–Introducing TWIST: Teleoperated Whole-Body Imitation System. We develop a humanoid teleoperation system to enable coordinated, versatile, whole-body movements, using a single neural network. This is our first step toward general-purpose robots. ๐ŸŒ https://t.co/ScrdX8ImNF
16
94
438
@jxwuyi
Yi Wu
8 months
๐ŸŽ‰ Milestone Release! AReaL-boba, our latest #RL system! https://t.co/xmZe676YIZ #AI โ€ข data/code/model ALL๐Ÿ”ฅ #OPENSOURCE โ€ข Full #SGLang & 1.5x faster on 7B RL โ€ข SOTA 7B math reasoning: 61.9 AIME24 & 48.3 AIME25 โ€ข 200-sample 32B tuning match QwQ on AIME24 @Alibaba_Qwen 1/3 ๐Ÿ‘‡
7
42
113
@mengdixu_
Mengdi Xu
8 months
Check out Yunfanโ€™s work on solving real household tasks! Very cool to see the robot not only grasping but also skillfully using different parts of its body to manipulate objects.๐Ÿšช๐Ÿค–๐Ÿ‘
@YunfanJiang
Yunfan Jiang
9 months
๐Ÿค– Ever wondered what robots need to truly help humans around the house? ๐Ÿก Introducing ๐—•๐—˜๐—›๐—”๐—ฉ๐—œ๐—ข๐—ฅ ๐—ฅ๐—ผ๐—ฏ๐—ผ๐˜ ๐—ฆ๐˜‚๐—ถ๐˜๐—ฒ (๐—•๐—ฅ๐—ฆ)โ€”a comprehensive framework for mastering mobile whole-body manipulation across diverse household tasks! ๐Ÿงน๐Ÿซง From taking out the trash to
2
1
31
@YunfanJiang
Yunfan Jiang
9 months
๐Ÿš€Two weeks ago, we hosted a welcome party for the newest member of our Stanford Vision and Learning Labโ€”a new robot! ๐Ÿค–โœจWatch as @drfeifei interacts with it in this fun video. Exciting release coming soon. Stay tuned! ๐Ÿ‘€๐ŸŽ‰
9
27
212
@mengdixu_
Mengdi Xu
10 months
Aligning object physical properties in sim with real is crucial to close sim2real gap, especially in nonprehensile manipulation. We propose CAPTURE which adapts simulator w/o gradient updates by treating real and sim rollouts as contexts. Please check out Xilunโ€™s ๐Ÿงตfor details!
@XilunZhang1999
XilunZhang
10 months
๐Ÿค– What if robots could adapt from simulation to reality on the fly, mastering tasks like scooping objects and playing table air hockey? ๐Ÿฅ„๐Ÿ“ Iโ€™m thrilled to share that our work, "Dynamics as Prompts: In-Context Learning for Sim-to-Real System Identification," has been accepted
1
2
40
@kevin_zakka
Kevin Zakka
10 months
The ultimate test of any physics simulator is its ability to deliver real-world results. With MuJoCo Playground, weโ€™ve combined the very best: MuJoCoโ€™s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of
37
186
909
@mengdixu_
Mengdi Xu
11 months
Excited to see that Genesis is officially released. Congratulations to the team! Looking forward to seeing how Genesis will advance robot policy learning, data generation, policy evaluation, and more!
@zhou_xian_
Zhou Xian
11 months
Everything you love about generative models โ€” now powered by real physics! Announcing the Genesis project โ€” after a 24-month large-scale research collaboration involving over 20 research labs โ€” a generative physics engine able to generate 4D dynamical worlds powered by a physics
1
0
39
@RuiChen_Rob
Rui Chen
1 year
Stay tuned!
@YifanSun98
Yifan Sun
1 year
[1/4] ๐ŸŒŸSneak Peek: SPARK in Action! ๐Ÿฆพ Previewing Safe Protective & Assistive Robot Kit (SPARK)โ€”a modular toolbox designed to enhance safety in humanoid autonomy and teleoperation. Safety isn't just a featureโ€”it's the foundation for humanoids to truly integrate into human life.
0
2
13
@drfeifei
Fei-Fei Li
1 year
Very excited to share with you what our teamย @theworldlabs has been up to! No matter how one theorizes the idea, it's hard to use words to describe the experience of interacting with 3D scenes generated by a photo or a sentence. Hope you enjoy this blog! ๐Ÿคฉโค๏ธโ€๐Ÿ”ฅ
@theworldlabs
World Labs
1 year
Weโ€™ve been busy building an AI system to generate 3D worlds from a single image. Check out some early results on our site, where you can interact with our scenes directly in the browser! https://t.co/ASD6ZHMwxI 1/n
78
264
2K
@elliottszwu
Elliott / Shangzhe Wu
1 year
I'm building a new research lab @Cambridge_Eng focusing on 4D computer vision and generative models. Interested in joining us as a PhD student? Apply to the Engineering program by Dec 3 ๐Ÿ—“๏ธ https://t.co/SDJEz2XiZp ChatGPT's "portrait of my current life"๐Ÿ‘‡ https://t.co/qcnSgqYMWr
4
43
218