Sign in

Lukas Muttenthaler

@lukasmut.bsky.social
437 followers 231 following 41 posts

Senior Researcher in AI4Health | Prev. Postdoc @eml-munich.bsky.social & SR at Google DeepMind | PhD in ML and NeuroAI from @tuberlin.bsky.social @bifold.berlin @mpicbs.bsky.social | Representations in 🧠 and 🤖 | #FirstGen 🌈 💻: lukasmut.github.io

PostsRepliesMedia
Lukas Muttenthaler @lukasmut.bsky.social · 22/07/2026
🔮At #ICML2026 we presented Attentive Multi-Layer Fusion for Vision Transformers, an efficient method for decoding information from multiple layers of ViTs for downstream task predictions. It’s almost as performant as full fine-tuning but only an inch more expensive than linear probing!
041
Reposted by Lukas Muttenthaler
ExplainableML @eml-munich.bsky.social · 03/07/2026
2/ Wed, Jul 8, 2026 • 10:30 AM – 12:15 PM KST · HALL A #2711 · POSTER Attentive Multi-Layer Fusion for Vision Transformers @lciernik.bsky.social , Marco Morik, @zeynepakata.bsky.social , @lukasmut.bsky.social Paper: arxiv.org/abs/2601.09322 Code: github.com/lciernik/att...
arxiv.org
Beyond the final layer: Attentive multilayer fusion for vision transformers
With the rise of large-scale foundation models, efficiently adapting them to downstream tasks remains a central challenge. Linear probing, which freezes the backbone and trains a lightweight head, is ...
111
Reposted by Lukas Muttenthaler
Tom Neuhäuser @tomneuhaeuser.bsky.social · 16/04/2026
Can context-sensitivity improve human–machine alignment in VFMs? We show that incorporating context improves modeling of human similarity decisions. 🎉Accepted to ReAlign ICLR! Now on @arxiv: arxiv.org/abs/2604.13883 Joint first-authored work with amazing Frieda Born. 1/5
182
Lukas Muttenthaler @lukasmut.bsky.social · 27/02/2026
Also signed :)
020
Reposted by Lukas Muttenthaler
BIFOLD Berlin Institute for the Foundations of Learning and Data @bifold.berlin · 20/02/2026
Researcher Spotlight: Dr. Lukas Muttenthaler BIFOLD PhD graduate Dr. Lukas Muttenthaler is pushing AI beyond raw performance, exploring #representational #alignment. More: t1p.de/pe5z5 @lukasmut.bsky.social @eml-munich.bsky.social @tuberlin.bsky.social @mpicbs.bsky.social @xai-berlin.bsky.social
181
Lukas Muttenthaler @lukasmut.bsky.social · 19/01/2026
🚨Go check out our most recent preprint on multilayer attention probing for Vision Transformers! It’s almost as performant as full fine-tuning while being similarly compute efficient as standard linear probing! Plus you get interpretable attention maps! More in 🧵👇🏼 Preprint: arxiv.org/abs/2601.09322
arxiv.org
Beyond the final layer: Attentive multilayer fusion for vision transformers
With the rise of large-scale foundation models, efficiently adapting them to downstream tasks remains a central challenge. Linear probing, which freezes the backbone and trains a lightweight head, is ...
041
Lukas Muttenthaler @lukasmut.bsky.social · 07/01/2026
Even though I am not part of the Re-Align organizer team anymore, I am still cheering for this workshop! I still (and will always) believe that Re-Align is a great place for connecting the fields of AI / ML, Neuroscience, and Cognitive Science :) Can highly recommend submitting! 🧠🤖
050
Reposted by Lukas Muttenthaler
Andrew Lampinen @lampinen.bsky.social · 23/12/2025
New post! Last week I shared why I thought cognitive (neuro)science hasn’t contributed as much as one might hope to the design of AI systems; this week I'm sharing my thoughts on how methods and principles from these fields *have* been useful in my work. infinitefaculty.substack.com/p/how-cognit...
infinitefaculty.substack.com
How cognitive science can contribute to AI: methods for understanding
#2 in a series on cognitive science and AI
2418
Lukas Muttenthaler @lukasmut.bsky.social · 07/12/2025
So well deserved! 💛
010
Reposted by Lukas Muttenthaler
Andrew Lampinen @lampinen.bsky.social · 12/11/2025
What aspects of human knowledge do vision models like CLIP fail to capture, and how can we improve them? We suggest models miss key global organization; aligning them makes them more robust. Check out LukasMuttenthaler's work, finally out (in Nature!?) www.nature.com/articles/s41... + our blog! 1/3
nature.com
Aligning machine and human visual representations across abstraction levels - Nature
Aligning foundation models with human judgments enables them to more accurately approximate human behaviour and uncertainty across various levels of visual abstraction, while additionally improving th...
28219
Reposted by Lukas Muttenthaler
Andrew Lampinen @lampinen.bsky.social · 12/11/2025
We summarize the ideas in a new DeepMind blog post: deepmind.google/blog/teachin... 2/3
deepmind.google
Teaching AI to See the World More Like Humans Do
Aligning AI vision models with human knowledge, improves their robustness and ability to generalize.
1142
Reposted by Lukas Muttenthaler
Max Planck Institute for Human Cognitive and Brain Sciences @mpicbs.bsky.social · 13/11/2025
When #AI “thinks” like us - New research out now in @nature.com from @lukasmut.bsky.social & @lampinen.bsky.social & @tuberlin.bsky.social & @mpib-berlin.bsky.social forms a bridge between #human and #machine representation. tinyurl.com/nhd4ycsh
tinyurl.com
When AI “thinks” like us
New research forms a bridge between human and machine representation
0165
Lukas Muttenthaler @lukasmut.bsky.social · 12/11/2025
🥳 I am incredibly humbled and grateful to share that our work, "Aligning machine and human visual representations across abstraction levels," has been published today in @nature.com ⬇️
0215
Reposted by Lukas Muttenthaler
Andrew Lampinen @lampinen.bsky.social · 29/10/2025
Pleased to share that our survey "Getting aligned on representational alignment" — on representational alignment across cognitive (neuro)science and machine learning — is now published in TMLR! openreview.net/forum?id=Hiq... Kudos to @sucholutsky.bsky.social @lukasmut.bsky.social for leading this!
openreview.net
Getting aligned on representational alignment
Biological and artificial information processing systems form representations of the world that they can use to categorize, reason, plan, navigate, and make decisions. How can we measure the...
04410
Reposted by Lukas Muttenthaler
Gregg Gonsalves @gregggonsalves.bsky.social · 13/09/2025
Though it’s traumatic for me to bring this up, I was targeted by a group allied with Turning Point USA in 2018, with the help of Tucker Carlson on Fox. My home address circulated online and I got death threats. 1/
8929361064
Reposted by Lukas Muttenthaler
Erin Grant @eringrant.me · 19/08/2025
I’m recruiting committee members for the Technical Program Committee at #CCN2026. Please apply if you want to help make submission, review & selection of contributed work (Extended Abstracts & Proceedings) more useful for everyone! 🌐 Helps to have: programming/communications/editorial experience.
31914
Lukas Muttenthaler @lukasmut.bsky.social · 14/08/2025
w/ @eringrant.me @dotadotadota.bsky.social @sucholutsky.bsky.social @siddsuresh97.bsky.social
020
Lukas Muttenthaler @lukasmut.bsky.social · 14/08/2025
Join us for the Re-Align^{3} community event & hackathon at #CCN2025 in room A2.11! There will be lots of 🌶️ discussions on 🧠🤖
1101
Reposted by Lukas Muttenthaler
Phillip Isola @phillipisola.bsky.social · 06/08/2025
Interesting reaction from ChatGPT to the HHS mRNA memo. It finds it so implausible that it thinks it's fake. From the perspective of a ~2024(?) trained model, 2025 policies are so absurd as to be unbelievable... chatgpt.com/share/689364...
35413
Reposted by Lukas Muttenthaler
Laure Ciernik @lciernik.bsky.social · 16/07/2025
🎉 Presenting at #ICML2025 tomorrow! Come and explore how representational similarities behave across datasets :) 📅 Thu Jul 17, 11 AM-1:30 PM PDT 📍 East Exhibition Hall A-B #E-2510 Huge thanks to @lorenzlinhardt.bsky.social, Marco Morik, Jonas Dippel, Simon Kornblith, and @lukasmut.bsky.social!
openreview.net
Objective drives the consistency of representational similarity...
The Platonic Representation Hypothesis claims that recent foundation models are converging to a shared representation space as a function of their downstream task performance, irrespective of the...
093
Reposted by Lukas Muttenthaler
Laure Ciernik @lciernik.bsky.social · 06/06/2025
If two models are more similar to each other than a third on ImageNet, will this hold for medical/satellite images? Our #icml2025 paper analyses how vision model similarities generalize across datasets, the factors that influence them, and their link to downstream task behavior. 🧵1/7
1244
Reposted by Lukas Muttenthaler
Martin Hebart @martinhebart.bsky.social · 23/06/2025
What makes humans similar or different to AI? In a paper out in @natmachintell.nature.com led by @florianmahner.bsky.social & @lukasmut.bsky.social, w/ Umut Güclü, we took a deep look at the factors underlying their representational alignment, with surprising results. www.nature.com/articles/s42...
nature.com
Dimensions underlying the representational alignment of deep neural networks with humans - Nature Machine Intelligence
An interpretability framework that compares how humans and deep neural networks process images has been presented. Their findings reveal that, unlike humans, deep neural networks focus more on visual ...
110436
Reposted by Lukas Muttenthaler
Laure Ciernik @lciernik.bsky.social · 06/06/2025
I am deeply grateful to @lorenzlinhardt.bsky.social, Marco Morik, Jonas Dippel, Simon Kornblith, and @lukasmut.bsky.social for their great work and support in this project! We also thank our collaborators, @bifold.berlin and HFA 7/7 📄Paper: arxiv.org/abs/2411.05561 💻Code: github.com/lciernik/sim...
arxiv.org
Objective drives the consistency of representational similarity across datasets
The Platonic Representation Hypothesis claims that recent foundation models are converging to a shared representation space as a function of their downstream task performance, irrespective of the obje...
063
Lukas Muttenthaler @lukasmut.bsky.social · 06/06/2025
🚨Have a look at our most recent work on the driving factors of representational similarities! This work will be presented at ICML 2025! See thread below 🧵
051
Lukas Muttenthaler @lukasmut.bsky.social · 22/05/2025
Please everyone gather and go outside, protest, march, do something! We cannot let science and academia be destroyed by the worst administration the US has ever had.
030
Reposted by Lukas Muttenthaler
Andrew Lampinen @lampinen.bsky.social · 02/05/2025
How do language models generalize from information they learn in-context vs. via finetuning? In arxiv.org/abs/2505.00661 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. 1/
arxiv.org
47722
Reposted by Lukas Muttenthaler
Dota Tianai Dong @dotadotadota.bsky.social · 28/04/2025
That’s a wrap for Re²-Align 🤖 🧠👤 at #ICLR2025 🦁🥳— thanks for being part of it! We’re probably just getting started… see you all at Re³-Align🌷😉
0272
Reposted by Lukas Muttenthaler
Siddharth Suresh @siddsuresh97.bsky.social · 28/04/2025
✨ 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
3165
Reposted by Lukas Muttenthaler
Barack Obama @barackobama.bsky.social · 15/04/2025
Harvard has set an example for other higher-ed institutions - rejecting an unlawful and ham-handed attempt to stifle academic freedom, while taking steps to make sure students can benefit from an environment of intellectual inquiry, rigorous debate and mutual respect. Let’s hope others follow suit.
15568916518162
Lukas Muttenthaler @lukasmut.bsky.social · 17/04/2025
The trump administration is the worst that could have happened.
010
Lukas Muttenthaler @lukasmut.bsky.social · 12/04/2025
This is terrible
000
Lukas Muttenthaler @lukasmut.bsky.social · 11/04/2025
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 at the Max Planck Institute for Human Cognitive and Brain Sciences www.cbs.mpg.de/cbs-coconut/...
cbs.mpg.de
video-muttenthaler
Lukas Muttenthaler (Google DeepMind, TU Berlin)
0113
Reposted by Lukas Muttenthaler
Eric Schulz @ericschulz.bsky.social · 11/04/2025
🚨 We're hiring! If you're excited about 🤖 ML/LLMs, 🧠 cognitive science, or 💭 computational psychiatry, come join us in Munich. Two fully funded PhDs @www.helmholtz-munich.de: tailored mentorship, international vibe, lots of room to grow. 📅 May 16th 🔗 hcai-munich.com/PhDHCAI.pdf
02410
Lukas Muttenthaler @lukasmut.bsky.social · 07/04/2025
Thank you, Rob! 🙏🏼
000
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
Thank you Desmond! Copenhagen definitely played a big part in preparing me for the PhD.
010
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
Thank so much, Andrew! You know you played a big part in all of this and I am grateful that I got to know you during my PhD and can call you both a mentor and a friend.
110
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
Oh wow, then congrats Dr. Greta!! How did I miss this? I probably saw it and said congrats but forgot about this lol. For some reason I was under the impression that you and I will defend at about the same time 🤔 Very happy for you!!
110
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
Thank you Greta! 🫶🏼 When’s your turn?
100
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
Thanks a lot for your kind words, Dota! 🙏🏼🫶🏼
000
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
I haven’t had an easy time recently (I struggled with mental health problems this past fall and winter) and I am so grateful for all the people who helped me find confidence and hope again to (successfully) finish this chapter. You all mean the world to me.
060
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
I‘d like to say thank you to everyone who supported me on this journey, to all the mentors I had, to all the people who believed in me! 🙏🏼
130
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
This past Friday I successfully defended my PhD 🎉 🎊👨🏻‍🎓 What a journey it was! 4.5 years of many ups and many downs. Can’t believe it’s over. I am still processing… Special thanks to my wonderful committee KR Müller, @martinhebart.bsky.social, @ericschulz.bsky.social, and @scychan.bsky.social!
7632
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
I haven’t had an easy time recently (I struggled with mental health problems this past fall and winter) and I am so grateful for all the people who helped me find confidence and hope again to (successfully) finish this chapter. You all mean the world to me.
000
Lukas Muttenthaler @lukasmut.bsky.social · 06/04/2025
I‘d like to say thank you to everyone who supported me on this journey, to all the mentors I had, to all the people who believed in me! 🙏🏼
100
Lukas Muttenthaler @lukasmut.bsky.social · 28/03/2025
Thank you for having me!
120
Reposted by Lukas Muttenthaler
CBS CoCoNUT @cbs-coconut.bsky.social · 28/03/2025
Join us for the talk by @lukasmut.bsky.social (Google Deep Mind, TU Berlin) on "Representational alignment of humans and machines for computer vision" Starting soon at 10 AM (CET), March 28 Find the Zoom Link here: www.cbs.mpg.de/en/cbs-coconut
cbs.mpg.de
CBS CoCoNUT
CBS CoCoNUT: A new interest group that comes together every two weeks to have an informal exchange about cognitive computational neuroscience.
032
Reposted by Lukas Muttenthaler
Martin Hebart @martinhebart.bsky.social · 17/11/2024
Training objective drives the consistency of representational similarity across datasets Laure Ciernik, Lorenz Linhardt, …, Simon Kornblith, Lukas Muttenthaler @lukasmut.bsky.social 👉 Indeed, people need to better appreciate the interaction of datasets and hypotheses arxiv.org/abs/2411.05561
arxiv.org
Training objective drives the consistency of representational similarity across datasets
The Platonic Representation Hypothesis claims that recent foundation models are converging to a shared representation space as a function of their downstream task performance, irrespective of the obje...
0216
Lukas Muttenthaler @lukasmut.bsky.social · 20/02/2025
I encourage all of my friends and fellow scientists in the US to go outside and march against Trump’s and Elon’s oligarchy. There’s nothing at this point that is more important than fighting for democracy! Democracy is in real danger and we here in Europa are concerned about the recent developments.
050
Reposted by Lukas Muttenthaler
Stephen Curry @scurry.bsky.social · 12/02/2025
Now almost 450 signatories to my open letter calling on the @royalsociety.org to stand up for its values & deal with the widespread concerns raised by Elon Musk's Fellowship. Please consider signing & sharing: forms.gle/miDciq35oxyw... List of signatories here: occamstypewriter.org/scurry/2025/...
forms.gle
Open letter to the President of the Royal Society – time to stand up for your values
If you wish to show your support for the letter below regarding apparent inaction by the Royal Society in the face of breaches of its code of conduct by Elon Musk FRS, please sign below. I invite anyo...
6272184
Reposted by Lukas Muttenthaler
Erin Grant @eringrant.me · 16/01/2025
Our representational alignment workshop returns to #ICLR2025! Submit your work on how ML/cogsci/neuro systems represent the world & what shapes these representations 💭🧠🤖 w/ @thisismyhat.bsky.social @dotadotadota.bsky.social, @sucholutsky.bsky.social @lukasmut.bsky.social @siddsuresh97.bsky.social
0209