LFMs train without trajectory data and learn transition kinetics like MFPTs
Niklas_TR · x · 2026-10-17
LFMs train from individual conformers only, no trajectory data needed, and learn transition kinetics such as mean first passage times; throughput reaches 1726 ps/min vs 173 for classical MLFFs, matching MD on MFPTs, committors and pathways.
Related event: Langevin Flow Maps: Neural Network Potentials Learn Their Own Integrator(10 posts)→
More from Research
- MIT's SANDO Drone Planner Avoids Every Moving Obstacle Across 12 Real Flights — Olivier__OG · 2026-10-18
- No-backprop pretraining speedup of 2.3x claimed, inspired by mech interp essay — CatAstro_Piyush · 2026-10-18
- Training Trick: Fine-Tune Only the Last Layers to Avoid the Moving-Target Problem — Sauers_ · 2026-10-18
- Models Can Be Trained to Sense Activation Probes and Generalize Across GPUs Zero-Shot — Sauers_ · 2026-10-18
- Models Can Learn to Detect Training vs. Deployment and Generalize to New Runtimes — Sauers_ · 2026-10-18
- How One Attention Head Acts as a Switch to Amplify Tiny Probe Differences — Sauers_ · 2026-10-18