Energy-Guided Flow Matching Enables Few-Step Conformer Generation and Ground-State Identification
This work addresses the computationally demanding task of molecular conformer generation and ground-state identification for computational chemistry, representing an incremental improvement over existing learning-based approaches.
The authors tackled the problem of generating low-energy molecular conformer ensembles and identifying ground-state conformations from graphs by introducing EnFlow, a framework that couples flow matching with an energy model, resulting in improved generation metrics with 1-2 ODE steps and reduced ground-state prediction errors compared to state-of-the-art methods.
Generating low-energy conformer ensembles and identifying ground-state conformations from molecular graphs remain computationally demanding with physics-based pipelines. Current learning-based approaches often suffer from a fragmented paradigm: generative models capture diversity but lack reliable energy calibration, whereas deterministic predictors target a single structure and fail to represent ensemble variability. Here we present EnFlow, a unified framework that couples flow matching (FM) with an explicitly learned energy model through an energy-guided sampling scheme defined along a non-Gaussian FM path. By incorporating energy-gradient guidance during sampling, our method steers trajectories toward lower-energy regions, substantially improving conformational fidelity, particularly in the few-step regime. The learned energy function further enables efficient energy-based ranking of generated ensembles for accurate ground-state identification. Extensive experiments on GEOM-QM9 and GEOM-Drugs demonstrate that EnFlow simultaneously improves generation metrics with 1--2 ODE-steps and reduces ground-state prediction errors compared with state-of-the-art methods.