Sign in

Günter Klambauer

@gklambauer.bsky.social
892 followers 176 following 134 posts

Deep Learning researcher | professor for Artificial Intelligence in Life Sciences | inventor of self-normalizing neural networks | ELLIS program Director

PostsRepliesMedia
Reposted by Günter Klambauer
Terence Tao @teorth.bsky.social · 11/09/2026
A group of 25 Fields Medalists, including myself, have made a joint declaration on Math and AI: mathandai.org . We welcome additional signatories. See also this article in the Economist announcing the declaration: www.economist.com/science-and-...
mathandai.org
Declaration — Math and AI
Read the declaration and add your name.
422073934
Günter Klambauer @gklambauer.bsky.social · 03/07/2026
G-RRM: Guiding Symbolic Solvers with Recurrent Reasoning Models Symbolic solver have to branch to check different choices -- recurrent reasoning models ( not LLMs!) can help them to speed up solving. BTW: Frontier AIs fail here! P: arxiv.org/abs/2607.02491
001
Günter Klambauer @gklambauer.bsky.social · 01/07/2026
LeWorld model becomes ADAPTIVE and meets MODEL-PREDICTIVE CONTROL AdaJEPA by Yann LeCun and colleagues performs actions, then checks the predicted latent state versus the observed and adapts at TEST TIME. Similar to meta-learning ... Paper: arxiv.org/abs/2606.32026
020
Günter Klambauer @gklambauer.bsky.social · 27/04/2026
Your AI model fails on new data? It will likely be due to the domain gap! Here is what you can do for biomedical AI models: P: arxiv.org/abs/2604.20824
110
Günter Klambauer @gklambauer.bsky.social · 17/04/2026
## Can LLMs Score Medical Diagnoses and Clinical Reasoning as well as Expert Panels? ## It appears so... "LLM Jury" has high correlation with the original score of a panel of human experts. Another human panel ("Re-score panel") does not correlate as much. P: arxiv.org/abs/2604.14892
010
Günter Klambauer @gklambauer.bsky.social · 06/03/2026
youtu.be/0qn2z9mD4Tc?...
youtu.be
From HRM to RRM: Recurrent Reasoning Models. Is the SE-RRM the Ultimate Sudoku Challenger?
YouTube video by AI Podcast Series. Byte Goose AI.
020
Günter Klambauer @gklambauer.bsky.social · 03/03/2026
Symbol-equivariant Recurrent Reasoning Models (SE-RRM) SE-RRM advances HRM and TRM -- guaranteed identical solutions for problems with permuted colors (ARC AGI) or digits (Sudoku). Coolest part: extrapolation to larger problem sizes!!! P: arxiv.org/abs/2603.02193 C: github.com/ml-jku/SE-RRM
051
Reposted by Günter Klambauer
Ö1 Digital Leben @oe1digitalleben.bsky.social · 06/02/2026
Mit #OpenClaw organisieren Nutzende ihr Leben. Der KI-Bot braucht dafür allerdings umfassenden Zugriff auf persönliche Daten. Und was hat es mit #Moltbook auf sich, wo sich diese Bots angeblich miteinander unterhalten? @gklambauer.bsky.social, Professor für KI an der JKU Linz, ordnet den Hype ein.
oe1.orf.at
Hype um KI-Agent | DO | 05 02 2026 | 17:55
Was steckt hinter OpenClaw und Moltbook?
031
Günter Klambauer @gklambauer.bsky.social · 27/01/2026
### NEW ENTRY IN THE AI IN DRUG DISCOVERY LEADERBOARD ### CheMeleon takes the lead as the **best pre-trained model** and gets the Bronze medal (rank 3) overall. It's an MPNN pre-trained to predict chemical descriptors. P: arxiv.org/abs/2506.15792 C: github.com/JacksonBurns...
021
Günter Klambauer @gklambauer.bsky.social · 15/01/2026
# AI in Drug discovery just BROKE THROUGH a wall # A newer AI model, ConGLUDe, as fast but much more accurate than DrugCLIP. Instead on just 40K structure-based data, ConGLUDe is trained on 100M datapoints from ligand-based data P: arxiv.org/abs/2601.09693
053
Günter Klambauer @gklambauer.bsky.social · 07/01/2026
A NEW ENTRY TO THE TOX21 LEADERBOARD: GROVER from TENCENT AI LAB obtains rank 5 on the Tox21 leaderboard. Best pre-trained model and clearly outperforms CHEMPROP. Original implementation: github.com/tencent-aila... Tox21 leaderboard on Hugging Face: huggingface.co/spaces/ml-jk...
011
Reposted by Günter Klambauer
hochreitersepp.bsky.social @hochreitersepp.bsky.social · 11/12/2025
xLSTM for Real-Time DNS Tunnel Detection: arxiv.org/abs/2512.09565 DNS-HyXNet = xLSTM for DNS tunnels. DNS-HyXNet has 99.99% accuracy, with F1-scores exceeding 99.96%, and per-sample detection latency of just 0.041 ms, confirming its scalability and real-time readiness. wow!
021
Reposted by Günter Klambauer
ELLIS Unit Linz & LIT AI Lab @litailab.bsky.social · 11/12/2025
Sepp Hochreiter, Head of LIT AI Lab, comments on the EurIPS conference in this FAZ guest article: He highlights Europe’s growing visibility in AI research and the role of ELLIS in strengthening the community. 🌍 www.faz.net/pro/digitalw...
faz.net
EurIPS wird zur Bühne für europäische KI-Talente
In Kopenhagen fand erstmals die EurIPS-Konferenz statt. Sie ist keine Gegenveranstaltung zur berühmten amerikanischen NeurIPS, sondern eine Bühne für erstklassige europäische KI-Forschung. Jetzt sind ...
021
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
Drug discovery benchmarks reveal a surprising result: the original 2017 SELU networks still dominate toxicity prediction (Tox21). P: arxiv.org/abs/2511.14744
arxiv.org
Measuring AI Progress in Drug Discovery: A Reproducible Leaderboard for the Tox21 Challenge
Deep learning's rise since the early 2010s has transformed fields like computer vision and natural language processing and strongly influenced biomedical research. For drug discovery specifically, a k...
000
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
GCN variants (e.g., GyralNet) increasingly replace ReLU with SELU for better convergence and robustness in noisy or shallow graph settings. P: arxiv.org/abs/2503.19823
arxiv.org
GyralNet Subnetwork Partitioning via Differentiable Spectral Modularity Optimization
Understanding the structural and functional organization of the human brain requires a detailed examination of cortical folding patterns, among which the three-hinge gyrus (3HG) has been identified as...
100
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
SELU made a comeback in RL: PPO-based systems like Pearl report smoother, more stable actor–critic updates when batchnorm is avoided. P: arxiv.org/abs/2506.01880
arxiv.org
Pearl: Automatic Code Optimization Using Deep Reinforcement Learning
Compilers are crucial in optimizing programs and accelerating their execution. However, optimizing programs automatically using compilers is not trivial. Recent work has attempted to use reinforcement...
100
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
The SELU-transformer resurged in NLP/tabular domains -- TabTranSELU even kicks out SwiGLU entirely. P: doi.org/10.54254/275...
doi.org
TabTranSELU: A transformer adaptation for solving tabular data
Tabular data are most prevalent datasets in real world, yet the integration of deep learning algorithms in tabular data often garners less attention despite their widespread utilization in other field...
100
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
Flowstate continued the trend: SELU activations bring stability in large-scale time-series models. P: arxiv.org/abs/2508.05287
arxiv.org
FlowState: Sampling Rate Invariant Time Series Forecasting
Foundation models (FMs) have transformed natural language processing, but their success has not yet translated to time series forecasting. Existing time series foundation models (TSFMs), often based o...
100
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
Time-series foundation models (FIM/FIM-ℓ, Flowstate) all quietly adopt SELU, and TiRex topped the GIFT Eval leaderboard this year. P: openreview.net/forum?id=NPS...
openreview.net
Zero-shot Imputation with Foundation Inference Models for Dynamical...
Dynamical systems governed by ordinary differential equations (ODEs) serve as models for a vast number of natural and social phenomena. In this work, we offer a fresh perspective on the classical...
100
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
Tiny SELU MLPs (2–3 layers, width 64) became the standard in Conditional Flow Matching, thanks to smoother derivative behavior than ReLU. P: arxiv.org/abs/2302.00482
arxiv.org
Improving and generalizing flow-based generative models with minibatch optimal transport
Continuous normalizing flows (CNFs) are an attractive generative modeling technique, but they have been held back by limitations in their simulation-based maximum likelihood training. We introduce the...
100
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
Normalization-Free Transformers rediscover controlled signal propagation — a core idea behind SNNs. P: arxiv.org/abs/2503.10622
arxiv.org
Transformers without Normalization
Normalization layers are ubiquitous in modern neural networks and have long been considered essential. This work demonstrates that Transformers without normalization can achieve the same or better per...
110
Günter Klambauer @gklambauer.bsky.social · 09/12/2025
The Great Comeback of Self-Normalizing Networks in 2025: It’s been a wild year in AI and for SNNs + SELU!! See my overview and some trends here: bioinf-jku.github.io/SNNs/
160
Reposted by Günter Klambauer
ELLIS Unit Linz & LIT AI Lab @litailab.bsky.social · 03/12/2025
Great news for AI research at JKU! Upper Austria invests €22.5M to strengthen AI & quantum science, supporting the ELLIS Unit Linz and the LIT AI Lab. This commitment boosts new projects, talent, collaborations & innovation made in Austria 🇦🇹 www.nachrichten.at/politik/land...
nachrichten.at
Land fördert die Kepler-Uni in den kommenden drei Jahren mit 22,5 Millionen Euro
LINZ. Dritte Fördervereinbarung sieht Schwerpunkte bei Künstlicher Intelligenz, Quantenwissenschaft und Internationalisierung vor.
021
Günter Klambauer @gklambauer.bsky.social · 21/11/2025
Thanks for more than 1000 visits to the leaderboard in the last 2 days! YOU THINK YOU HAVE A BETTER MODEL?? GREAT! Training models on the Tox21 Challenge dataset can be done on a laptop-GPU. Submit your model to the leaderboard! Click SUBMIT here: huggingface.co/spaces/ml-jk...
huggingface.co
Tox21 Leaderboard - a Hugging Face Space by ml-jku
This application allows you to manage and view a leaderboard. You can submit scores or data to update the leaderboard, which refreshes daily at 2 AM.
020
Günter Klambauer @gklambauer.bsky.social · 19/11/2025
AI is hitting a wall in drug discovery! Here is the wall you were all looking for... huggingface.co/spaces/ml-jk...
010
Günter Klambauer @gklambauer.bsky.social · 19/11/2025
Yes, ECFP plus some other descriptors.. should be in the model-cards and the paper ..
120
Günter Klambauer @gklambauer.bsky.social · 19/11/2025
You have access to the model predictions via API call. Just type: curl -X POST ml-jku-tox21-gin-classifier.hf.space/predict -H "Content-Type: application/json" -d '{"smiles": ["CCO", "c1ccccc1"]}' in your shell.
030
Günter Klambauer @gklambauer.bsky.social · 19/11/2025
Measuring AI Progress in Drug Discovery - A NEW LEADERBOARD IN TOWN 2015-2025: turns out that there's hardly any improvement. AI bubble? GPT is at 70% for this task, whereas the best methods get close to 85%. Leaderboard: huggingface.co/spaces/ml-jk... P: arxiv.org/abs/2511.14744
3127
Günter Klambauer @gklambauer.bsky.social · 17/11/2025
List of accepted contributions now online: moleculediscovery.github.io/workshop2025/
000
Günter Klambauer @gklambauer.bsky.social · 14/11/2025
Roughly 50% of the tickets already gone...
000
Günter Klambauer @gklambauer.bsky.social · 14/11/2025
Indeed.. :)
000
Günter Klambauer @gklambauer.bsky.social · 13/11/2025
How many errors do you find in the chemical structure above?
000
Reposted by Günter Klambauer
Mario Zechner @mariozechner.at · 13/11/2025
Hab mir einen Mahrer Tracker gebaut. Spannend zu sehen, wie schnell Print-Medien auf ihren Online-Portalen auf Dinge wie ein ZIB 2 Interview reagieren und wie sich die Inhalte dabei unterscheiden.
24513
Günter Klambauer @gklambauer.bsky.social · 13/11/2025
ELLIS ML4Molecules Workshop 2025: Registration for virtual participation now open!!! Only 500 tickets available which usually are gone fast!! Register now: www.eventbrite.com/e/ellis-mach...
eventbrite.com
ELLIS Machine Learning for Molecules Workshop 2025 - Virtual participation
Workshop on generative Models, LLMs, and the Future of Molecular AI. For details, see: https://moleculediscovery.github.io/workshop2025/
254
Günter Klambauer @gklambauer.bsky.social · 13/11/2025
Speaker 4 at ELLIS ML4Molecules 2025: Nadine Schneider Early works on ML methods for chemical reactions. Introduced the FS-Mol datasets for few-shot learning for molecules. Director of the ELLIS ML for Molecules Discovery program. Join the workshop: moleculediscovery.github.io/workshop2025/
020
Günter Klambauer @gklambauer.bsky.social · 12/11/2025
Invited speaker 3 at #ML4Molecules2025: Daniel Probst ( @skepteis.bsky.social ). He is known for his works on exploring chemical spaces, fingerprints, language models,... and some unconventional ideas like learning on gzip compressed data. Join: moleculediscovery.github.io/workshop2025/
092
Günter Klambauer @gklambauer.bsky.social · 10/11/2025
@marwinsegler.bsky.social will talk about "Deep Learning for Molecules: The First Decade" at #ML4Molecules2025 In pioneering work, he introduced a "small LLM" for generating SMILES strings back in 2016/17. Chemical synthesis prediction with MCTS; ... Join: moleculediscovery.github.io/workshop2025/
030
Günter Klambauer @gklambauer.bsky.social · 30/10/2025
I feel Bluesky is really dead...
120
Reposted by Günter Klambauer
Richard Kueng @richardkueng.bsky.social · 21/10/2025
(I/III) We're excited to announce a new tenure track opening! The position is called 'quantum informatics' and is affiliated with our QUICK group within the CS+AI division at @jku.at 🇦🇹. Application deadline is November 30th, 2025: www.jku.at/en/the-jku/w...
12817
Günter Klambauer @gklambauer.bsky.social · 21/10/2025
Celebrating 4,000 citations! Thanks to everyone who successfully used self-normalizing networks!!!
060
Günter Klambauer @gklambauer.bsky.social · 10/10/2025
Deadline extended to October 15th!
020
Günter Klambauer @gklambauer.bsky.social · 07/10/2025
PAPER/ABSTRACT DEADLINE ALREADY END OF THIS WEEK! ELLIS Machine Learning for Molecules workshop: moleculediscovery.github.io/workshop2025/ DON'T MISS THE DEADLINE: short papers or extended abstracts welcome!
010
Günter Klambauer @gklambauer.bsky.social · 29/09/2025
ELLIS ML4Molecules workshop 2025 ** Call for papers** is out: moleculediscovery.github.io/workshop2025/
011
Günter Klambauer @gklambauer.bsky.social · 24/09/2025
Is Bluesky the actual dead Internet??
110
Günter Klambauer @gklambauer.bsky.social · 23/09/2025
It's happening again!!! ML4Molecules workshop 2025. within the #ELLIS Unconference, preceding #EurIPS. More infos: moleculediscovery.github.io/workshop2025/
095
Günter Klambauer @gklambauer.bsky.social · 17/09/2025
Happy that IBM uses proper MLPs, i.e. self-normalizing networks, in their Flowstate model! ;) P: arxiv.org/abs/2508.05287
010
Günter Klambauer @gklambauer.bsky.social · 16/09/2025
Another win for xLSTM
000
Günter Klambauer @gklambauer.bsky.social · 16/09/2025
It just struck me that the ancient Greeks chose their twelve Olympian gods with gender parity..
010
Günter Klambauer @gklambauer.bsky.social · 08/09/2025
We Austrians are fully aware that we are a democratic country now, but we still sneak in pictures of our monarchs... :)
010
Günter Klambauer @gklambauer.bsky.social · 13/08/2025
Controversial posts are downranked on Blue sky???? Oh c'mon...
010