I am very proud and excited about this project! We are learning a general particle simulator directly from videos end-to-end. Check it out 👇 #Simulation #Gausiansplatting #ML

EWRL is happening in Tübingen this year! Checkout our 🦋 posts. When? September 17-19, 2025. bsky.app/profile/gmarti…

☁️💻Coming soon: AMLEC – the Atmospheric Machine Learning Emulation Challenge. 📅Stay tuned for details. Launching April 21st. 🔗If you want to participate and to submit your results you'll need an account ➡️huggingface.co/datasets/isp-u… ℹ️ECML-PKDD 2025: ecmlpkdd.org/2025/submissio…

Now featuring: a full upper-body biomechanical model that learned to reach (SCONE + Hyfydy)

Excited to present our paper “Unifying Causal Representation Learning with the Invariance Principle” at #ICLR2025 in Singapore! Joint work with DarioRancati, Riccardo Cadei , MarcoFumero, and Francesco Locatello.

Congratulation to Jonas Frey for a successful defense on "Learning Perception and Navigation for autonomous robots in the wild"! D-MAVT, ETH Zurich ETH Zurich thanks Maurice Fallon Georg Martius NASA JPL

It was a pleasure to work with you!

Consider submitting your talk proposals to EWRL! 🐣 More info in 🦋 @ewrl18.bsky.social and the website …p-on-reinforcement-learning.github.io/ewrl18/

Enabling robots to learn from humans would be a game changer! But humans may manipulate objects in a way that is impossible for robots. To cross this embodiment gap, Human2Sim2Robot follows this key insight: Don't imitate a demonstration but instead use it to guide RL.

How to represent dynamic real-world data both consistently and efficiently, while reflecting the compositional object-centric structure of the world? Contrast your slots! ...with our new SlotContrast method(🚀#CVPR2025 Oral🚀)! 🌐website: slotcontrast.github.io 🧵🧵🧵 1/n

🚨 “The Diffusion Duality” is out! International Conference on Minority Languages ⚡️ Few-step generation in discrete diffusion language models by exploiting the underlying Gaussian diffusion. 🦾Beats AR on 3/7 zero-shot likelihood benchmarks. 📄 Paper: arxiv.org/abs/2506.10892 💻 Code: github.com/s-sahoo/duo 🧠

Getting from Tübingen participating in the Huggingface LeRobot hackathon! #Huggingface #Robotarm #hackathon #PhysicalAI #TuebingenAI huggingface.co/spaces/LeRobot…

All forms of intelligence co-emerged with a body, except AI We're building a #future where AI evolves as your lifelike digital twin to assist your needs across health, sports, daily life, creativity, & beyond... myolab.ai ➡️ Preview your first #HumanEmbodiedAI

Lesen! Als Klimaforscher kann ich dem, was Christian Stöcker schreibt, nur voll zustimmen. Es ist zum Haareraufen. spiegel.de/wissenschaft/m…

✨Introducing SENSEI✨ We bring semantically meaningful exploration to model-based RL using VLMs. With intrinsic rewards for novel yet useful behaviors, SENSEI showcases strong exploration in MiniHack, Pokémon Red & Robodesk. Accepted at ICML 2025🎉 Joint work with Christian Gumbsch 🧵

🚨 The NeurIPS Conference #MyoChallenge2025 environments are now LIVE! 📦 Install with: `pip install MyoSuite>=2.9.0` Get a head start before the competition officially opens! ⚽ Penalty kicks & 🏓 Table tennis rallies Start training today 👉 lnkd.in/dkzCiybT

When multiple tasks need improvements, fine-tuning a generalist policy becomes tricky. How do we allocate a demonstration budget across a set of tasks of varied difficulty and familiarity? We are presenting a possible solution at ICML on Wednesday! (1/3)

Zero-shot imitation from just a single sparse demonstration is hard. Goal-conditioned methods tend to “greedily" move from one state to the next and lose the big picture. We're presenting an alternative approach on Tuesday at #ICML2025. (1/3)