Flow-Matching Based Refiner for Molecular Conformer Generation
This addresses a foundational problem in drug discovery by refining conformer generation, though it is incremental as it builds on existing denoising methods.
The paper tackles error accumulation in molecular conformer generation by proposing a flow-matching refiner that initializes from mixed-quality outputs and reschedules noise scales to bypass low-SNR phases, improving sample quality on GEOM-QM9 and GEOM-Drugs benchmarks with fewer denoising steps while preserving diversity.
Low-energy molecular conformers generation (MCG) is a foundational yet challenging problem in drug discovery. Denoising-based methods include diffusion and flow-matching methods that learn mappings from a simple base distribution to the molecular conformer distribution. However, these approaches often suffer from error accumulation during sampling, especially in the low SNR steps, which are hard to train. To address these challenges, we propose a flow-matching refiner for the MCG task. The proposed method initializes sampling from mixed-quality outputs produced by upstream denoising models and reschedules the noise scale to bypass the low-SNR phase, thereby improving sample quality. On the GEOM-QM9 and GEOM-Drugs benchmark datasets, the generator-refiner pipeline improves quality with fewer total denoising steps while preserving diversity.