{"slug": "equifilm-charge-conditioned-equivariant-force-fields-via-feature-wise-linear", "title": "EquiFiLM: Charge-Conditioned Equivariant Force Fields via Feature-wise Linear Modulation", "summary": "Researchers introduced EquiFiLM, a lightweight extension that adds continuous external conditioning to equivariant foundation machine learning force fields via per-layer Feature-wise Linear Modulation. Applied to charged liquid water using the MACE-MatPES backbone, the method achieved a 3.1× reduction in force RMSE and a 61× reduction in energy RMSE over a baseline, requiring only thousands of DFT-labeled frames versus the millions needed for a charge-aware foundation trained from scratch.", "body_md": "arXiv:2607.05559v1 Announce Type: new\nAbstract: Foundation machine learning force fields (MLFFs) such as MACE-MP-0 and UMA cover broad chemical space at near density functional theory (DFT) accuracy. However, they assume equilibrium ground-state physics and do not natively handle externally induced changes to the electronic state, such as charging, applied fields, or electronic excitation, which limits their use for driven processes such as photoexcitation and charge injection. We propose EquiFiLM, a lightweight extension that adds continuous external conditioning to any equivariant foundation MLFF via a per-layer Feature-wise Linear Modulation (FiLM) block, learning externally driven changes to the potential energy surface from minimal training data. The block modulates only scalar channels and preserves E(3)-equivariance exactly. We demonstrate the recipe on charged liquid water with the foundation model MACE-MatPES as the backbone, yielding E-MACE. On the four training charges, E-MACE delivers a $3.1\\times$ reduction in force RMSE ($21.3$ to $6.96$ meV/$\\mathring{A}$) and a $61\\times$ reduction in per-atom energy RMSE ($6.1$ to $0.1$ meV/atom) over a baseline without EquiFiLM trained on the same data, at indistinguishable inference cost. Across seven held-out interpolation and extrapolation charges, force RMSE stays within $18-61$ meV/$\\mathring{A}$ and energy RMSE within $0.7-5.4$ meV/atom. The model runs stable molecular dynamics across the full range tested and predicts the charge-dependent first-shell response of the reduced pair distribution function probed by ultrafast electron diffraction. Adding this conditioning axis to the foundation requires only a few thousand DFT-labeled frames, against the $\\approx 10^8$ structures of a charge-aware foundation trained from scratch. The recipe is backbone- and conditioning-agnostic: it applies without architectural change to any equivariant MLFF with scalar interaction-layer channels.", "url": "https://wpnews.pro/news/equifilm-charge-conditioned-equivariant-force-fields-via-feature-wise-linear", "canonical_source": "https://arxiv.org/abs/2607.05559", "published_at": "2026-07-08 04:00:00+00:00", "updated_at": "2026-07-08 04:17:02.263770+00:00", "lang": "en", "topics": ["machine-learning", "ai-research", "ai-infrastructure"], "entities": ["EquiFiLM", "MACE-MP-0", "UMA", "MACE-MatPES", "E-MACE", "DFT"], "alternates": {"html": "https://wpnews.pro/news/equifilm-charge-conditioned-equivariant-force-fields-via-feature-wise-linear", "markdown": "https://wpnews.pro/news/equifilm-charge-conditioned-equivariant-force-fields-via-feature-wise-linear.md", "text": "https://wpnews.pro/news/equifilm-charge-conditioned-equivariant-force-fields-via-feature-wise-linear.txt", "jsonld": "https://wpnews.pro/news/equifilm-charge-conditioned-equivariant-force-fields-via-feature-wise-linear.jsonld"}}