DINOv3 just released — I'm excited to see its application in robotics
21.08.2025 07:22
👍 2
🔁 0
💬 0
📌 0
DINOv3 just released — I'm excited to see its application in robotics
SCMs often assume overly-dense causal graphs in dynamic settings 👉⚽🥎, since any object interaction is a potential causal edge, making them hard to scale. In joint work with @turanorujlu.bsky.social @cgumbsch.bsky.social & Martin Butz we propose a new Causal Process Model to tackle this.
Thread👇
Sergey Levine was just presenting in the Exploration in AI @ #ICML2025 and promoted that exploration needs to be grounded, and that VLMs are a good source ;-) Check our paper below
👇
I am in Vancouver at ICML, and tomorrow I will present our newest paper "Partially Observable Reinforcement Learning with Memory Traces". We argue that eligibility traces are more effective than sliding windows as a memory mechanism for RL in POMDPs. 🧵
✨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 @cgumbsch.bsky.social
🧵
News!⚡
We’re pleased to announce a new Call for Contributed Talks to bring novel and diverse viewpoints to EWRL2025.
Early-career researchers are especially encouraged to submit proposals and share their work with the community.
Full details: euro-workshop-on-reinforcement-learning.github.io/ewrl18/
🎤 Meet the speakers!
First up: Amy Zhang (amyzhang.github.io), assistant professor at UT Austin and head of the Machine Intelligence through Decision-making and Interaction (MIDI) Lab. Her group works on theory and algorithms for sequential decision-making problems.
Time to submit your paper! EWRL submissions are open!💫
Check out the Call for Papers here: euro-workshop-on-reinforcement-learning.github.io/ewrl18/
Introducing 3DGSim🧩— an end-to-end 3D physics simulator trained only on multi-view videos. It achieves spatial & temporal consistency w/o ground truth 3D info or heavy inductive biases— enabling scalability & generalization🚀
Kudos to Mikel + @andregeist.bsky.social
www.youtube.com/watch?v=3Ar3...
Hiring announcement: ELLIS Institute Tübingen is looking for ML Researchers & Engineers for Open-Source AI Tutoring (m/f/d). The image features a white background with bold black text and the colorful ELLIS logo at the bottom.
🚀 We’re hiring! Join Bernhard Schölkopf & me at @ellisinsttue.bsky.social to push the frontier of #AI in education!
We’re building cutting-edge, open-source AI tutoring models for high-quality, adaptive learning for all pupils with support from the Hector Foundation.
👉 forms.gle/sxvXbJhZSccr...
Wir Demokratielaien sind der Mob, der rülpsend und polternd jeden Tag eine neue Sau durchs Dorf jagt. Im Bundestag, Herr Friedrich Merz, brauchen wir entschlossene Besonnenheit und keine Abgeordneten, die ihre Mistgabeln und Fackeln mit zur Arbeit bringen und damit den Faschisten Feuer geben.
The call for workshops is out. Now may be a bit busy but the deadline is on March 7th so there’s time :)
Send us your great ideas for 1-day workshops at RLC!
image demonstrating rules about axes, handedness, and direction of rotation in 3D space, where you can use your thumb as the axis, and the curl of your fingers as the positive rotation direction
handedness can be a little hard to wrap your head around - it's a very elusive concept as it's mostly perceptual/visual rather than being a property of the numbers/math itself
notably though, you can use it to figure out positive rotation directions, given an arbitrary axis!
BO with Variational Last Layers as Surrogate
📚 New Paper with collaborators from DTU, Vector, & Google DeepMind
A neural net-based approach to BO that performs well in both classic, small-scale problems, and can efficiently scale far beyond GP surrogate models.
Visit our poster @ NeurIPS Bayesian decision-making workshop today!
More info👇
