There are some collaborations that are just fun and this was one of them! Big thank you to Anna Meier and Alessandra Moretti and everyone else involved!🫀

A new method from Johannes Wirth [email protected] Celia P. Martinez-Jimenez Meier Lab - MiBioEng @ HPC enables cost effective spatial transcriptomics on multiple tissues in parallel. PioneerCampus @HelmholtzMunich

📺New video: Curvature for Graph Learning📺 Presenting, among other things, current work with (i) Josh Southern, Jeremy Wayland, Michael Bronstein and (ii) Corinna Coupette and Sebastian Dalleiger. Very excited about this research direction! youtube.com/watch?v=-V0Jpd…

Great talk Bastian, thanks for uploading!

Happy to announce our new paper on Deterministic Uncertainty Methods TrustML-(Un)Limited #ICLR2023 ! We dissect how the design of training schemes, architecture, and prior can significantly impact feature collapse and uncertainty performance! w/ C. Zhang and Stephan Günnemann

1/9 CPAis finally published. It can predict single-cell responses to combinatorial perturbation (drugs, CRISPR). This is a joint collab between @meta and Fabian Theis Helmholtz Munich | @HelmholtzMunich. Read the thread to understand the LEGO analogy! embopress.org/doi/full/10.15…

If you are interested in ML potentials and/or uncertainty estimation come and visit our presentation of "Uncertainty Estimation for Molecules: Desiderata and Methods" this week at #ICML! Tue, 11am, #415 Joint work with Nicholas Gao Bertrand Charpentier Amine Ketata Stephan Günnemann 1/2

Do you care about uncertainty quantification? Do you like guarantees? Check out our #ICML2023 paper where we bring conformal prediction to Graph Neural Networks. tldr: Instead of a single label return a *set* that is guaranteed to contain the true label. Set size = uncertainty.

Is binarization of scATAC-seq data really necessary? The conclusion from our analysis is that a quantitative treatment is in fact beneficial. Now out in Nature Methods! gagneurlab Fabian Theis nature.com/articles/s4159… Many additions since the preprint 👇(1/n)

Thrilled to present our paper "(Provable) Adversarial Robustness for Group Equivariant Tasks" at #NeurIPS2023! Joint work with great collaborators @YanScholten Stephan Günnemann! Paper: arxiv.org/abs/2312.02708 Poster #800 Thursday 11 am - 13 pm #NeurIPS

Get to know #mcml junior member Leon Hetzel 💡 He developed an algorithm called chemCPA designed to make drug discovery more effective. Read more (in German): mcml.ai/news/2024-03-2… This article was written by Leonie Fischer from Deutsche Journalistenschule!

Mixed Models with Multiple Instance Learning (MixMIL) received an Oral & Outstanding Student Paper award at AISTATS Conference last week! 🏆 MixMIL enables accurate & interpretable patient label prediction from single-cell data by adding attention to GLMMs.#singlecell #MachineLearning

Presenting at #ICML2024: Fragment-Biases for Molecular GNNs 🧪 Tue, 23.07: Oral session 1F @ 11:00 Poster #105 @ 11:30 🔑 Fragment-Biased GNNs outperform others, match Transformers with better generalization & linear cost! 🤝 With N. Kemper, Leon Hetzel , Johanna Sommer , Stephan Günnemann

11 exciting news: Our group has 10 papers at #NeurIPS2024 (incl. 1 oral + 2 spotlights) 📃🎓. And as of October 1st, I am on entrepreneurial leave 🚀. Re papers: Congrats to all co-authors. Amazing work! go.tum.de/689644 Re startup: We are hiring! pruna.ai

Three papers NeurIPS Conference 2024 🎉 An efficient adv. training algorithm for LLMs arxiv.org/abs/2405.15589 Unlearned LLMs are not safe against adv. attacks arxiv.org/abs/2402.09063 Scaling robustness of Lipschitz-1 networks arxiv.org/abs/2305.10388 Happy to chat in Vancouver!

Unified Guidance for Geometry-Conditioned Molecular Generation • UniGuide introduces a unified framework for geometry-conditioned molecular generation using diffusion models. It provides a flexible method for controlled molecular design, eliminating the need for additional

It’s rainy in Vancouver, poster hall is closed but we are ready 🙌 👉 Come and talk to us and learn about UniGuide at Poster#2600 (East) UniGuide is a new framework for molecular diffusion models that enables flexible geometric conditioning across tasks—no retraining required

Congrats to my amazing PhD students: We have 9 papers accepted at #ICLR2025. Reliability, AI4Science, graphs, LLMs, and more (go.tum.de/936150). And if you follow the recent discussions about AI efficiency, you might like our blog and webinars (pruna.ai).

Finding good resources on efficient AI is harder than it should be. We're fixing that! 🚀 Check out our new github.com/PrunaAI/awesom… repo —a curated hub of the best tools, papers, and techniques to make AI faster, smaller, and cheaper.