Jin Cheng

Doctoral Student at @crl_ethz @ETH, working on loco-manipulation, robot learning with @StelianCoros and @GuanyaShi. Cells interlinked within cells, interlinked.

[Danfei Xu @danfei_xu]

Monday thoughts:

Since deep learning, there are two “root-level” paradigm shifts in robot learning: Sim2Real for locomotion and Behavior Cloning (BC) for manipulation. 1/

[Fan Shi ✈️RSS 2026 @fanshi_robot]

Can learned deformable manipulation policies improve at inference time — without retraining?

Yes: with real2sim2real + sampling-based MPC.

🟣 Novel physics-grounded deformable simulation with fast, accurate, GPU-parallel real-time rollouts
🟣 RGB-based real2sim state estimation
🟣 Policy-agnostic refinement that boosts inference-time success rates
🟣 Runs on a single RTX 4090 GPU

Project page: https://t.co/AasT7ITBET
More details in the thread below 👇

[Jin Cheng @catachiii]

@ShuoYangAIR Cannot wait! Huge congrats!

[Genesis AI @gs_ai_]

Humanoid robots don't need to look human.

Meet Eno, our first general-purpose robot.

Not a machine pretending to be human, but intelligence given a body.

At Genesis, we’re building a future where robots don’t feel cold or distant, but capable, calm, and ready to help.

Available Q4 this year.

[dar @radbackwards]

Turns out Gundam designers make for great humanoid designers

[Jin Cheng @catachiii]

I still remember the first course I worked as a teaching assistant was Dynamic Programming and Optimal Control. Most of the contents are from Prof. Bertsekas’ book with the same name. Young and great minds from ETH are always fascinated by him. Not only academically inspiring, he is also a great friend in photography.

You will always be remembered, Prof. Bertsekas. RIP.

[Yi Ma @YiMaTweets]

Professor Dimitri Bertsekas will be missed by many and many more to come… RIP.🙏🙏🙏

[Jin Cheng @catachiii]

@YiMaTweets :(

[Jin Cheng @catachiii]

@KKawaharazuka Not NERV? :D

[Núria Armengol @NriaArmengol2]

🐊Tomorrow, our student Yuanchen Yuan is presenting CAIMAN, “Causal Action Influence Detection for Sample-efficient Loco-Manipulation” at @ieee_ras_icra.

Proud to have contributed alongside Yuanchen, Jin Cheng @catachii and Stelian Coros.
Go talk to them ⭐️

[Jin Cheng @catachiii]

A massive thank you to all my brilliant collaborators across @ETH_en, @UniversityKorea, @rai_inst, @ZHAW, and @CMU_Robotics! 🙌

If you're at the conference, let's talk robotics or grab a coffee!☕

[Jin Cheng @catachiii]

Vienna is calling! 🚀

Pumped to be heading to Austria next week for #ICRA2026. It’s been an intensive year, and I'm honored to present 4 papers pushing the boundaries in legged robotics, MPC, and quadruped motion retargeting. 🐕🦾

Here is where to find me: 👇🧵

[Jin Cheng @catachiii]

📌 THURSDAY
⏰ 15:00 - 16:30
📄 RAMBO: RL-Augmented Model-Based Whole-Body Control for Loco-Manipulation (Poster ThI2I.19)

[Michael Pan @mpan31415]

Touch is important for dexterous manipulation, but hard to use in sim-to-real RL.

Binary contact transfers well, but loses rich contact information; raw tactile readings preserve information, but create a large sim-to-real gap.

[Computational Robotics Lab @crl_ethz]

We’re heading to Vienna! 🇦🇹🤖

The Computational Robotics Lab (CRL) @ETH_en is thrilled to present 5 papers at #ICRA2026 next week! We're bringing our latest work on legged loco-manipulation, deformable objects, MPC, and RL.

Drop by our interactive sessions! 👇 Thread 🧵

[Jin Cheng @catachiii]

@weishuaizeng Where's the arxiv 😍

[Jin Cheng @catachiii]

@rlfromlux Pierre, this is so awesome! Looking forward to what’s coming!

[Guanya Shi @GuanyaShi]

My lab at CMU @LeCARLab is hiring a postdoc! (Vibe-made the poster via @ChatGPTapp)

[Yuxuan Mu @YuxuanMu16173]

Can we build a standalone, modular, and reusable naturalness reward for training motor controllers?

#SMP is a step toward that vision. Once SMP has been trained on a motion dataset, the priors can be reused to train new controllers to perform diverse tasks while adhering to the behaviors in the dataset, without original dataset or retraining.

🔥 Excited to share our latest work, SMP: Score-Matching Motion Priors, accepted to @siggraph

Webpage: https://t.co/Pz4yFAg1wo
Code: https://t.co/rZPp5b5GPD
Paper: https://t.co/K0z1oQkdFZ
Video: https://t.co/gPkyQCqNWz

[Jin Cheng @catachiii]

@zhaomingxie wooooow