🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

De novo protein design is great, but nature has millions of proteins- why not repurpose them? Introducing Raygun, a new approach to protein design. It allows you to miniaturize, magnify or modify any protein. We synthesized miniaturized variants of eGFP and mCherry! 1/

Cats are invariant under SO(3) transformations! 😼

BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”

Check out our latest work proposing a novel way to model mRNA sequences with language models by encoding biological hierarchical priors. Improves downstream property prediction and sequence generation tasks by 8% on avg. Joint work with amazing Artem Moskalev 🕊️ and Mehdi Yazdani

🌟 Excited to announce that we’ll be bringing our track, "AI in Computational Biology: From Molecules to Medicine", to #AMLDEPFL2025! If you work on 1️⃣ Integrative Multi-Omics or 2️⃣ Predictive & Generative Modeling for structural biology, consider submitting your poster by Nov 5

In our latest work on the impact of using geometry for RNA property prediction with Junjie Xu and Artem Moskalev 🕊️, we tackle data scarce RNA domain and show that geometry is useful in data and label scarce scenario but reliance on geometry for noisy and OOD scenarios hurts

Completely agree! Also the review quality is much lower in general. Seems like making reviewing mandatory for all authors for papers with certain number of submissions does not necessarily lead to better review quality

Excited to see our two papers on mRNA language modeling and RNA property prediction accepted at #ICLR2025! Big thanks to amazing Artem Moskalev 🕊️ and our two interns Mehdi Yazdani and Junjie Xu for the nice work! Check out the details below! #ML #RNA #ICLR2025

As one of the people who popularized the field of diffusion models, I am excited to share something that might be the “beginning of the end” of it. IMM has a single stable training stage, a single objective, and a single network — all are what make diffusion so popular today.

How is the tick box exercise of just acknowledging rebuttal without responding at all supposed to improve reviewer-author discussion ICML Conference ? The authors get no rebuttal feedback and reviewers have no accountability as long as they tick acknowledgement button #icml2025

Come to our poster tomorrow if you’re at #ICLR25 and interested in AI/ML for drug discovery!

Into AI/ML for drug discovery and at #ICLR2025 in Singapore 🇸🇬! Come meet us at our posters. Main Conf •HELM: Thurs 10–12:30 | Hall 3 + Hall 2B #6 •Beyond Sequence: Sat 10–12:30 | Hall 3 + Hall 2B #5 AI4NA (Orals) •InfoSEM: Mon •HARMONY: Mon Let’s connect!

arxiv.org/abs/2412.06264

if you want to know more, see * lecture note: arxiv.org/abs/2505.03861 * syllabus: docs.google.com/document/d/1On…

Check out our recent work on how to build geometric equivarince for Hyena models enabling long context structural modeling with sub-quadratic complexity! Accepted at #icml2025 as spotlight. Amazing work by the talented Artem Moskalev 🕊️ and the team!