Jin Chen

phd student on EmbodiedAI

[Aran Komatsuzaki @arankomatsuzaki]

PingpongDuo

Humanoid table tennis now has doubles, while I'm still single.

proj: https://t.co/JONxxxDWPu
abs: https://t.co/Wm1PJmYYTq

[Jin Chen @nsabsw]

@Selen7005717917 seedance is all you need. 😁

[Jin Chen @nsabsw]

🔬 Sharing EgoHumanoid-V2, my recent work at Tongyi Lab with our collaborators.

Human-only task demonstrations → learned whole-body coordination → zero-shot deployment on real humanoids.

🌱 A step toward scalable WBI foundation models.

🔗 https://t.co/61Iu4EpcES

[Jin Chen @nsabsw]

🧠 After Astra, I keep asking: what remains fundamentally hard in embodied AI?

My bet is on instinctive behaviors—the whole-body coordination that comes naturally to humans.

🤖 From human bodily intelligence to Whole-Body Intelligence (WBI).

Our exploration 👇

[Jin Chen @nsabsw]

@OpenAI CRAZY

[OpenAI @OpenAI]

We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.

The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.

The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.

[Jin Chen @nsabsw]

@berryxia great

[RoboPapers @RoboPapers]

Full episode dropping soon!

Geeking out with @idomihs on EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration https://t.co/NooyUn4vXN

Co-hosted by @micoolcho @DJiafei

[Jin Chen @nsabsw]

@Kongkongda5882 中国的科研教育还需要沉淀

[Jin Chen @nsabsw]

@PKU1898 祝贺！

[Simons Foundation @SimonsFdn]

During today's #ICM2026 opening ceremonies, the IMU announced the 2026 Fields Medals recipients:

@UChicago's Yu Deng, @stonybrooku's John Pardon, @UofT's Jacob Tsimerman, and Hong Wang of @nyuniversity and @Institut_IHES.

Read more: https://t.co/oJARAI8I8j

[Jin Chen @nsabsw]

@SimonsFdn @CSProfKGD @UChicago @stonybrooku @UofT @nyuniversity @Institut_IHES Congratulations!

[Techniahqrobot | humanoid robots @techniahqrobot]

This robot is getting way too good at this.

It can be even faster!
A master makes the move
Just enough to score

From : @Accelerated Evolution

[Jin Chen @nsabsw]

@mikekalilmfg it even uses fake hands!!!

[Jin Chen @nsabsw]

@NVIDIARobotics @LightwheelAI @AdventHealth Nice.

[Lord Bebo @MyLordBebo]

🇮🇳 AI DATA RACE: Workers in Indian factories started to wear cameras on their heads, to film their motions so robots can be trained on these videos.

Big robot companies will train their humanoid robots, on movement data from Indian sweatshops … Wild

1/

[Haochen Tian @HCTian713]

SimScale is selected as CVPR 2026 Oral 🎉🥳

Code, simulation data, and checkpoints have been fully open-sourced.

📷Home: https://t.co/ejTgqZ3NbW
📷Paper: https://t.co/euJW2MOtOE
📷Github: https://t.co/AxuTMvqVGk

If you find something useful, please give it a star⭐

[OpenDriveLab @OpenDriveLab]

https://t.co/7uGphEXxjk

[Jiazhi Yang @jiazhi_yang2024]

RISE (1/N)

Before diving into technical details, let's first see what we've achieved on challenging manipulation.

RISE yields absolute performance increases of:
+45% in Backpack Packing
+35% in Box Closing
+35% in Dynamic Brick Sorting
compared to a strong offline RL method (RECAP, re-implementated on pi_0.5, which is an amazing work by @physical_int)

These improvements highlight the effectiveness of leveraging a proper world model for policy optimization for contact-rich manipulation.

[OpenDriveLab @OpenDriveLab]

【1/5】❓Why do Vision-Language Navigation (VLN) agents need "step-by-step" language instructions? No.❌ We only want to give a intent. 📍

Meet SparseVideoNav: a new paradigm that uses video generation to navigate like a human. Tell the destination. Imagine the path. Reach the goal. 🌄🤖

Paper: https://t.co/Eq9nxbGskQ
Page：https://t.co/FSPJ9Qu0KD
Github：https://t.co/y1lDJ0y6wr