“What does a good chain-of-thought look like for molecules?" Excited to share our work, “Chain-of-Thoughts for Molecular Understanding,” at the AIDrugX workshop in #NeurIPS2024! 📍 Location: West Meeting Room 109, 110 🕒 Time: 4:30 PM arxiv.org/abs/2410.05610

I decided to start tweeting about our work! We have five papers (MOF structure prediction / transition path sampling, GFlowNet exploration / synthetic pathway GFlowNet / molecule structure prediction) at #ICLR2025. See you at Singapore! Paper details in the thread.

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)

Today we release TorchSim, a next-generation PyTorch-native atomistic simulation engine for the MLIP era. Our hope is that this open-source package ignites a materials simulation revolution.

Starting my day with a 5 min vibe coding session. Behold, the marginal probability path with a Gaussian conditional path used in Flow Matching. Input: Screenshot of LaTeX equations and instructions Output: videos

Excited to be in Singapore next week for #ICLR2025! 🇸🇬 DM if you want to chat about geometric deep learning and/or generative models on manifolds, or just to enjoy some nice specialty coffee ☕️ I don't look like Barbie but my poster does 🪩😎

Symmetry groups in materials 😀

[Notable Reviewer @ ICLR25] I'm honored to be selected as a Notable Reviewer this year after reviewing 4 papers! It was a exciting experience to engage with cutting-edge research in ML force fields (MLFF), DFT, molecular dynamics, and equivariant spherical representations. Thanks

The supervision signal in AI4Science is so crisp that we can solve very complicated problems almost without any data or RL! In this project, we train a model to solve the Schrödinger equation for different molecular conformations using Density Functional Theory (DFT) In the

Excited to share our #ICML2025 Spotlight paper "On the Guidance of Flow Matching", a framework of exact energy guidance in flow matching models. This work theoretically studies how to achieve controlled generation in the powerful generative model, flow matching, and proposes an

NeurIPS rebuttal deadline is around the corner 😬 I’m not an expert, but thought I’d drop my two cents on how to write a good rebuttal, especially for folks writing their first few. Hope this helps someone! 🧵👇 (And please chime in with your own tips; let’s crowdsource the

🔬 I will present our work “Structural Reasoning Improves Molecular Understanding of LLM” at #ACL2025 today! 📍 Poster #292 | Hall X4 🗓️ July 30, 2025 – 11:00–12:30 Looking forward to engaging discussions and feedback. See you in Hall X4, Poster 292! 🚀 #LLM #AI4Science