Karsten Kreis (@karsten_kreis) 's Twitter Profile
Karsten Kreis

@karsten_kreis

Principal Research Scientist at @NVIDIA | Former Physicist | Deep Generative Learning.

Opinions are my own.

ID: 1402885929937915905

linkhttps://karstenkreis.github.io/ calendar_today10-06-2021 07:10:57

368 Tweet

2,2K Followers

621 Following

Arash Vahdat (@arashvahdat) 's Twitter Profile Photo

Our new #ICLR2025 oral shows how you can improve designability and diversity of generated proteins by guiding their generation with 3D blobs. This allows scientists and protein designers to create and manipulate proteins with simple structural primitives.

Karsten Kreis (@karsten_kreis) 's Twitter Profile Photo

🔥Large-scale protein structure generation? Flow Matching? Fold class guidance? Big transformers? State-of-the-art benchmark performance?🔥 Very excited for this session on Friday, with the amazing Tomas Geffner, & hosted by Michael Bronstein, Bruno Correia! Hope to see you there! 😀

VantAI (@vant_ai) 's Twitter Profile Photo

Our talk with Karsten Kreis and Tomas Geffner on Proteina—Scaling Flow-based Protein Structure Generative Models—is now available on YouTube: youtu.be/-Qlc4dNj4e8?si…

Our talk with <a href="/karsten_kreis/">Karsten Kreis</a> and <a href="/tomasgeffner/">Tomas Geffner</a> on Proteina—Scaling Flow-based Protein Structure Generative Models—is now available on YouTube: youtu.be/-Qlc4dNj4e8?si…
VantAI (@vant_ai) 's Twitter Profile Photo

Announcing Neo-1: the world’s most advanced atomistic foundation model, unifying structure prediction and all-atom de novo generation for the first time - to decode and design the structure of life 🧵(1/10)

Woody Ahern (@woodyahern) 's Twitter Profile Photo

New enzymes can unlock chemistry we never had access to before. Here, we introduce RFdiffusion2 (RFD2), a generative model that makes significant strides in de novo enzyme design. Preprint: biorxiv.org/content/10.110… Code: coming soon Animation credit: x.com/ichaydon (1/n)

Arash Vahdat (@arashvahdat) 's Twitter Profile Photo

📢 EquiVDM: Equivariant Video Diffusion Models with Temporally Consistent Noise What's the role of equivariance in video diffusion models? and how can warped noise help with it? Is sampling from equivariant video models any easier? Project: research.nvidia.com/labs/genair/eq… w/ Chao Liu.

Jason Yim (@json_yim) 's Twitter Profile Photo

I'll be at ICLR. Come check out our generative modeling work! Reach out if you want to chat. Proteina: x.com/karsten_kreis/… Protcomposer: x.com/HannesStaerk/s… Generator matching: x.com/peholderrieth/…

Jason Yim (@json_yim) 's Twitter Profile Photo

I'll be at the ICLR GEMBio Workshop workshop presenting latent and structure diffusion for protein backbone generation. Come by to talk all things latent for biology. openreview.net/forum?id=Ek7Hs… arxiv.org/abs/2504.09374

Arash Vahdat (@arashvahdat) 's Twitter Profile Photo

🔥📔 This week at #ICLR2025, our fundamental generative AI research team (GenAIR) is (co-)presenting 11 papers, 6 of which were developed or led primarily by our team members. Below, I am listing our main papers with a one-sentence summary.

Kieran Didi (@didikieran) 's Twitter Profile Photo

Excited to present my first paper officially as a PhD student now as an ICLR Oral this week! Super fun work with the GenAIR team at NVIDIA. Talk: Fr 10:54 - 11:06 (Oral Session 3B, Garnet 213-215) Poster: Fr 15:00-17:30 (Hall 3 + Hall 2B #5) Come by the poster/reach out to chat

Anthony Costa (@anthonycosta) 's Twitter Profile Photo

Congratulations to Peter Holderrieth Michael Albergo @ICLR2025 and Tommi Jaakkola for winning the best paper award for their work entitled "LEAPS: A discrete neural sampler via locally equivariant networks" at this year's Frontiers in Probabilistic Inference workshop #ICLR2025!

Congratulations to <a href="/peholderrieth/">Peter Holderrieth</a> <a href="/msalbergo/">Michael Albergo @ICLR2025</a> and Tommi Jaakkola for winning the best paper award for their work entitled "LEAPS: A discrete neural sampler via locally equivariant networks" at this year's Frontiers in Probabilistic Inference workshop #ICLR2025!
Fabian Falck (@fabianfalck) 's Twitter Profile Photo

Is low-to-high frequency generation in diffusion models aka. 'approximate spectral autoregression' a necessity for generation performance? 📃 Blog: fabianfalck.com/posts/spectral… 📜 Paper: arxiv.org/abs/2505.11278

Austin Cheng (@auhcheng) 's Twitter Profile Photo

Excited to share Quetzal, a simple but scalable model for building 3D molecules atom-by-atom. 🐉 Named after Quetzalcoatl, the Aztec god of creation We equip a standard causal transformer with a per-atom diffusion MLP to model the continuous 3D position of the next atom. [1/3]