Langevin Flow Maps: ML force fields that learn their own integrator run 10x faster
Niklas_TR · x · 2026-10-17
A new paper (arXiv:2610.05998) introduces Langevin Flow Maps, extending machine-learned force fields to also learn the stochastic Langevin integrator. Key points:
- Standard MD requires femtosecond timesteps, making long-timescale transition dynamics hard to estimate.
- The method enables large-timestep MD while recovering accurate dynamical properties.
- Runs an order of magnitude faster than current ML force fields and requires no trajectory data.
- Trained on a diverse molecular dataset, showing a path toward transferable potentials.
Code and data are open-sourced by McCallum, Rindtorff, Tong, and Foster.
Related event: Langevin Flow Maps: Neural Network Potentials Learn Their Own Integrator(10 posts)→
More from Research
- Opus-assisted open design: a $31k low-field head MRI sketched in hours — DreamFly_13 · 2026-10-18
- semiconductor.diy: open-source guides to make real chips in a garden shed — jwt0625 · 2026-10-18
- Work in progress: geometric structure of GPG projections as Markov chain coarsening — michaelchchoi · 2026-10-18
- Backprop-free pretraining gets 2.3x speedup as Noise method crushes Dust record on 8 GPUs — inductionheads · 2026-10-18
- q3dviewer: open-source Qt+OpenGL tool for painter-like point cloud editing — 4310sy · 2026-10-18
- Smarter Every Day's Destin to defend PhD dissertation on impact fusion — CatAstro_Piyush · 2026-10-18