Differentiable hybrid force fields support scalable autonomous electrolyte discovery

Xintian Wang, Junmin Chen, Zhuoying Zhu, Peichen Zhong · 2026

Read Differentiable hybrid force fields support scalable autonomous electrolyte discovery by Xintian Wang, Junmin Chen, Zhuoying Zhu, Peichen Zhong (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-07979.txt.

Description

Wang and three colleagues argue that differentiable hybrid force fields resolve a trilemma in autonomous electrolyte discovery. A discovery engine must screen fast, predict properties well, and refine itself online from new data. Classical empirical force fields run fast but lean on error cancellation, while standard machine-learning potentials cost too much compute. Hybrid fields fuse physics-based functional shapes with neural-network short-range fixes. Grounded in energy decomposition analysis, models like PhyNEO-Electrolyte and ByteFF-Pol generalize zero-shot to bulk phases, reaching tens of nanoseconds per day.

Catalog record

CREATORXintian Wang, Junmin Chen, Zhuoying Zhu, Peichen Zhong
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33286
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.07979

Browse the full library · About the Tower of Babel