MolGuidance: Advanced Guidance Strategies for Conditional Molecular Generation with Flow Matching

TL;DR

This paper introduces advanced guidance strategies for conditional molecular generation using flow matching, achieving state-of-the-art property alignment and structural validity, with a comprehensive comparison of guidance methods.

Contribution

It integrates multiple guidance techniques into a molecular generation framework, proposing a hybrid approach that separately guides molecular modalities and optimizes guidance scales.

Findings

Achieves new state-of-the-art property alignment on QM9 and QMe14S datasets.

Generated molecules exhibit high structural validity.

Systematic comparison of guidance methods reveals their strengths and limitations.

Abstract

Key objectives in conditional molecular generation include ensuring chemical validity, aligning generated molecules with target properties, promoting structural diversity, and enabling efficient sampling for discovery. Recent advances in computer vision introduced a range of new guidance strategies for generative models, many of which can be adapted to support these goals. In this work, we integrate state-of-the-art guidance methods -- including classifier-free guidance, autoguidance, and model guidance -- in a leading molecule generation framework built on an SE(3)-equivariant flow matching process. We propose a hybrid guidance strategy that separately guides continuous and discrete molecular modalities -- operating on velocity fields and predicted logits, respectively -- while jointly optimizing their guidance scales via Bayesian optimization. Our implementation, benchmarked on the QM9 and QMe14S datasets, achieves new state-of-the-art performance in property alignment for de novo molecular generation. The generated molecules also exhibit high structural validity. Furthermore, we systematically compare the strengths and limitations of various guidance methods, offering insights into their broader applicability.