Gen-COMPAS: diffusion model proposes, physics decides for rare molecular transitions
bravo_abad · x · 2026-09-17
- Rare molecular events like protein folding or membrane transport take microseconds or longer, and most simulation time is spent waiting for something interesting.
- Gen-COMPAS has a diffusion model propose plausible intermediate structures where a transition might occur; molecular dynamics under the true Hamiltonian then tests and refines them.
- The loop: generate candidates → test with physics → keep promising transition states → retrain → repeat. A learned committor focuses the search on structures with roughly equal probability of falling into either end state.
- Demonstrated on the Trp-cage protein, exemplifying the "generative model proposes, physics decides" AI-for-Science paradigm.
More from Research
- Dream-RSI: Google DeepMind paper makes the research process itself self-improving — McDonaghMatthew · 2026-09-17
- Dream-RSI from Google DeepMind turns AI experiment logs into reusable self-improvement testbeds — McDonaghMatthew · 2026-09-17
- MIT's FloatForm robot boats self-assemble into bridges on water, published in Nature — lukas_m_ziegler · 2026-09-17
- DeepMind open-sources a fruit fly that learned to fly, 59-dim action space on a laptop — anthara_ai · 2026-09-17
- XPeng unveils XPACE, a unified embodied world model powering its IRON humanoid — CyberRobooo · 2026-09-17
- Max-pooling attribution maps model outputs back to input words in new sparse-model release — antoine_chaffin · 2026-09-17