Elign: Equivariant Diffusion Model Alignment from Foundational Machine Learning Force Fields

TL;DR

Elign is a novel framework that enhances 3D molecular conformation generation by integrating physics-based guidance into diffusion models, using pretrained MLFFs and reinforcement learning to ensure stability and accuracy without costly runtime evaluations.

Contribution

Elign introduces a post-training method that replaces expensive quantum evaluations with pretrained MLFFs and shifts physical guidance to training via reinforcement learning, enabling fast, accurate molecular conformation sampling.

Findings

Generates conformations with lower DFT energies and forces.

Improves stability of generated molecular structures.

Maintains fast inference speed without runtime energy evaluations.

Abstract

Generative models for 3D molecular conformations must respect Euclidean symmetries and concentrate probability mass on thermodynamically favorable, mechanically stable structures. However, E(3)-equivariant diffusion models often reproduce biases from semi-empirical training data rather than capturing the equilibrium distribution of a high-fidelity Hamiltonian. While physics-based guidance can correct this, it faces two computational bottlenecks: expensive quantum-chemical evaluations (e.g., DFT) and the need to repeat such queries at every sampling step. We present Elign, a post-training framework that amortizes both costs. First, we replace expensive DFT evaluations with a faster, pretrained foundational machine-learning force field (MLFF) to provide physical signals. Second, we eliminate repeated run-time queries by shifting physical steering to the training phase. To achieve the second amortization, we formulate reverse diffusion as a reinforcement learning problem and introduce Force--Energy Disentangled Group Relative Policy Optimization (FED-GRPO) to fine-tune the denoising policy. FED-GRPO includes a potential-based energy reward and a force-based stability reward, which are optimized and group-normalized independently. Experiments show that Elign generates conformations with lower gold-standard DFT energies and forces, while improving stability. Crucially, inference remains as fast as unguided sampling, since no energy evaluations are required during generation.