⚠️Beware: Your AI assistant could be hijacked just by encountering a malicious image online! Our latest research exposes critical security risks in AI assistants. An attacker can hijack them by simply posting an image on social media and waiting for it to be captured. [1/6] 🧵

Yesterday, we shared the details on our xLSTM 7B architecture. Now, let's go one level deeper🧑‍🔧 We introduce ⚡️Tiled Flash Linear Attention (TFLA), ⚡️ A new kernel algorithm for the mLSTM and other Linear Attention variants with Gating. We find TFLA is really fast! 🧵(1/11)

1/11 Excited to present our latest work "Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics" at #ICLR2025 on Fri 25 Apr at 10 am! #CombinatorialOptimization #StatisticalPhysics #DiffusionModels

🧠➕🔢 Research automation in action! We generated a new paper: primality test using circulant matrix eigenvalue structures—in just 24 hours. From math concept → formal paper → working implementation, our neurosymbolic platform compressed what would typically take months.

Could not feel prouder of the team at ELLIS Alicante !! Two accepted papers ICML Conference ! ✨ Both papers are creative, innovative and the result of fruitful collabs thanks ELIAS #ELSA grants and ELLIS PhD program! Congrats everyone!!🥳

This is the AI doom scenario I am actually afraid of, not the BS propagated by attention seekers. The real risk is a gradual and silent influence on society and its institutions (prob. so slow we won't notice for a long time), not AI suddenly taking over. Great work!

Excited to share that 2 of our papers on efficient inference with #xLSTM are accepted at #ICML25. A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks (arxiv.org/abs/2410.22391) and xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference:

Need to predict bioactivity 🧪 but only have limited data ❌? Try our interactive app for prompting MHNfs — a state-of-the-art model for few-shot molecule–property prediction. No coding or training needed. 🚀 📄 Paper: pubs.acs.org/doi/10.1021/ac… 🖥️ App: huggingface.co/spaces/ml-jku/…

Happy to introduce 🔥LaM-SLidE🔥! We show how trajectories of spatial dynamical systems can be modeled in latent space by --> leveraging IDENTIFIERS. 📚Paper: arxiv.org/abs/2502.12128 💻Code: github.com/ml-jku/LaM-SLi… 📝Blog: ml-jku.github.io/LaM-SLidE/ 1/n

1. We often observe power laws between loss and compute: loss = a * flops ^ b + c 2. Models are rapidly becoming more efficient, i.e. use less compute to reach the same loss But: which innovations actually change the exponent in the power law (b) vs change only the constant (a)?

There were so many great replies to this thread, let's do a Part 2! For scaling laws between loss and compute, where loss = a * flops ^ b + c, which factors change primarily the constant (a) and which factors can actually change the exponent (b)? x.com/_katieeverett/…

We’re excited to introduce TiRex — a pre-trained time series forecasting model based on an xLSTM architecture.

Attention!! Our TiRex time series model, built on xLSTM, is topping all major international leaderboards. A European-developed model is leading the field—significantly ahead of U.S. competitors like Amazon, Datadog, Salesforce, and Google, as well as Chinese models from Alibaba.

Ever wondered how linear RNNs like #mLSTM (#xLSTM) or #Mamba can be extended to multiple dimensions? Check out "pLSTM: parallelizable Linear Source Transition Mark networks". #pLSTM works on sequences, images, (directed acyclic) graphs. Paper link: arxiv.org/abs/2506.11997

We release AB-UPT, a novel method to scale neural surrogates to CFD meshes beyond 100 million of mesh cells. AB-UPT is extensively tested on the largest publicly available datasets. 📄 arxiv.org/abs/2502.09692 🤗 huggingface.co/EmmiAI/AB-UPT 💻 github.com/Emmi-AI/AB-UPT

The day has come! After intense months of work, involving a thousand stakeholders, the #AIAct #CodeofPractice for #GPAI models has been published! Thank you to all the participants in the process and especially the (vice)-chairs Yoshua Bengio Marietje Schaake Matthias Samwald

General relativity 🤝 neural fields This simulation of a black hole is coming from our neural networks 🚀 We introduce Einstein Fields, a compact NN representation for 4D numerical relativity. EinFields are designed to handle the tensorial properties of GR and its derivatives.

Uncertainty quantification (UQ) is key for safe, reliable LLMs... but are we evaluating it correctly? 🚨 Our ACL2025 paper finds a hidden flaw: if both UQ methods and correctness metrics are biased by the same factor (e.g., response length), evaluations get systematically skewed