Lukas Muttenthaler

Really excited to share what we’ve been working on for the past 12 months during my time @GoogleDeepMind! We came up with an approach that can distill the hierarchical structure of human conceptual knowledge into vision foundation models via a surrogate teacher model. More below!

How do language models generalize from information they learn in-context vs. via finetuning? We show that in-context learning can generalize more flexibly, illustrating key differences in the inductive biases of these modes of learning — and ways to improve finetuning. Thread: 1/

Elon Musk might have intentionally mischaracterized the AfD as centrist for several reasons. 1. Political alignment: He may share AfD's Eurosceptic and anti-immigration views, aiming to legitimize them. 2. Influencing public opinion: His high-profile status could normalize…

Beautiful work by @AndrewLampinen and @scychan_brains as ever

How can we circumvent data scarcity in the time series domain? We propose to leverage pretrained ViTs (e.g., CLIP, DINOv2) for time series classification and outperform time series foundation models (TSFMs). 📄 Preprint: arxiv.org/abs/2506.08641 💻 Code: github.com/ExplainableML/…

We find self-supervised & language-aligned ViTs scored highest on CSS — even matching humans. Supervised models are not even close. Surprisingly, high ImageNet accuracy does not guarantee high configural shape score! 5/15

"On the Ability of Deep Networks to Learn Symmetries from Data: A Neural Kernel Theory" now accepted at JMLR! 🥳 🔗arxiv.org/abs/2412.11521 We thank the reviewers for expert suggestions which allowed us to substantially improve the work and writing. See ⬇️ for more info and 🧵

Independence Day is a reminder that America is not the project of any one person. The single most powerful word in our democracy is the word ‘We.’ ‘We The People.’ ‘We Shall Overcome.’ ‘Yes We Can.’ America is owned by no one. It belongs to all citizens. And at this moment in…

More than 16 million Americans are at risk of losing their health care because Republicans in Congress are rushing to pass a bill that would cut federal funding for Medicaid and weaken the Affordable Care Act. If the House passes this bill, it will increase costs and hurt…

This is the syllabus of the course @geoffreyhinton and I taught in 1998 at the Gatsby Unit (just after it was founded). Notice anything?

Does #AI perceive and make sense of the world the same way humans do? @florianmahner, @lukas_mut & @martin_hebart @jlugiessen investigated whether AI recognizes objects similarly to humans and published their findings @NatMachIntell: tinyurl.com/2krukhzf

🚨Have a look at our most recent work on the driving factors for representational similarities of neural network models! This work will be presented at @icmlconf in just about 1 month by @lciernik and Marco!

🎉 Update: This work got accepted to #icml2025!! Huge thanks to my amazing co-authors @LorenzLinhardt, Marco Morik, @jdppel, @skornblith, and @lukas_mut for their great work and to all collaborators! 🙏 📄 Paper: arxiv.org/abs/2411.05561 💻 Code: github.com/lciernik/simil… 🧵1/3

Here's a nice "proof without words": The sum of the squares of several positive values can never be bigger than the square of their sum. This picture helps make sense of how ℓ₁ and ℓ₂ norms regularize and sparsify solutions (resp.). [1/n]

I think Elon Musk should be expelled from the British Royal Society. Not because he peddles conspiracy theories and makes Nazi salutes, but because of the huge damage he is doing to scientific institutions in the US. Now let's see if he really believes in free speech.

🤔 Curious about what "representational alignment" actually means? 🧩 Check out our freshly updated preprint! 📝✨ We've incorporated exciting findings and contributions from the previous edition into this comprehensive update! x.com/sucholutsky/st…

🔄✨ Come join us at the Second edition of the Re-Align workshop @iclr_conf! 🚀🧠 The workshop explores the fascinating question of how artificial and biological systems align in their representations of the world. #ReAlign #ICLR2025

Excited to share that our paper "Bridging the human–AI knowledge gap through concept discovery and transfer in AlphaZero" is now out in PNAS! With @weballergy, @banburismus_, @demishassabis, @ulrichpaquet, @_beenkim 🎉 📄 doi.org/10.1073/pnas.2…

“No government—regardless of which party is in power—should dictate what private universities can teach, whom they can admit and hire, and which areas of study and inquiry they can pursue.” - President Alan Garber hrvd.me/GarberRespond3…

Here’s an extended version of my PhD defense talk about our work on *representational alignment between humans 🧠 and machines 🤖 for computer vision* that I gave two weeks ago @MPI_CBS cbs.mpg.de/cbs-coconut/lu…