SA-LapNet fixes spin exactly in neural-network quantum Monte Carlo, breaking the purity-energy trade-off
bravo_abad · x · 2026-10-25
Li and colleagues tackle a key failure mode in neural-network quantum Monte Carlo: networks accidentally mixing unwanted spin states into calculations. The usual fix — penalizing mixing during training — forces a trade-off, improving spin purity while worsening energy accuracy on the tested iron–sulfur cluster. Their SA-LapNet approach instead fixes the required total spin exactly by mathematical construction of the wave function, letting the network learn only spatial electron relationships. Results beat AFQMC across 15 small systems while breaking the spin-penalty Pareto frontier, at lower per-step cost.
More from Research
- Safety researcher: verification should go all the way down to Maxwell's equations, not stop at assembly — JacquesThibs · 2026-10-25
- Local 35B authored 800 math problems under cloud-agent supervision — authoring erred 5x more than cold-solving — Individual_Page9676 · 2026-10-25
- XYBench: Strongest LLMs Answer Literally and Detect Under 63% of User Misconceptions — MaartenSap · 2026-10-25
- Multi-domain training mix boosts supervised hallucination detectors 7.25% on out-of-distribution QA — MaartenSap · 2026-10-25
- EVALUESTEER: best reward models fail to align with user values over 25% of the time — MaartenSap · 2026-10-25
- Ouvia study: only about half of real-world speech translation interactions rated usable — MaartenSap · 2026-10-25