Michael Plainer

Excited to share our latest preprint: 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗛𝗮𝗺𝗶𝗹𝘁𝗼𝗻𝗶𝗮𝗻 𝗙𝗹𝗼𝘄 𝗠𝗮𝗽𝘀: 𝗠𝗲𝗮𝗻 𝗙𝗹𝗼𝘄 𝗖𝗼𝗻𝘀𝗶𝘀𝘁𝗲𝗻𝗰𝘆 𝗳𝗼𝗿 𝗟𝗮𝗿𝗴𝗲-𝗧𝗶𝗺𝗲𝘀𝘁𝗲𝗽 𝗠𝗼𝗹𝗲𝗰𝘂𝗹𝗮𝗿 𝗗𝘆𝗻𝗮𝗺𝗶𝗰𝘀 🎉

@guforosso2 @leonklein26 Unfortunately, there is no free lunch. For the system shown, our model trained about 7 days. I think this approach is especially promising once it can be transferred. We also trained a model that can simulate all dipeptides, but this is not feasible for larger systems yet ...

@leonklein26 Efficient ? Number of machine cycle o tflops ?

(1/n) Can diffusion models simulate molecular dynamics instead of generating independent samples? In our NeurIPS2025 paper, we train energy-based diffusion models that can do both: - Generate independent samples - Learn the underlying potential 𝑼 🧵👇 arxiv.org/abs/2506.17139

@trendradar_app @leonklein26 @DoctorYev We compare our results with the ground-truth simulation from the fast-folding protein paper and note close alignment between the two. What I find interesting is that the FES can look a bit oversmoothed, possibly due to the "simple" architecture we use that does not use any priors

@leonklein26 @DoctorYev Interesting idea. Curious how the results compare to traditional methods. Any surprises in the dynamics you saw?

@giannis_daras @leonklein26 Thank you for sharing this! This indeed looks very related, and we will have a closer look 🙏

@leonklein26 Great work! I think you will be interested in our prior (and related) work on this: Consistent Diffusion models. arxiv.org/abs/2302.09057 We also talk a lot about violations in the FP, and we propose one way to mitigate this.

Fantastic work by Michael Plainer and friends. Energy-based diffusion models to ensure that the denoising distribution equals exp(-u(x)), with the energy u(x). A keystone for connecting molecular dynamics, statistical mechanics and generative AI.