Property-Guided Molecular Generation and Optimization via Latent Flows

Alexander Arjun Lobo, Urvi Awasthi, Leonid Zhukov

Keywords: Molecular generation, Inverse molecular design, Latent space optimization, Flow matching, Multi-objective optimization

TL;DR: MoltenFlow is a latent-space framework that combines property-organized representations with an unconditional flow-matching prior and surrogate-guided inference to enable stable, multi-objective molecular generation and optimization.

Abstract:

Molecular discovery is increasingly framed as an inverse design problem: identifying molecular structures that satisfy desired property profiles under feasibility constraints. While recent generative models provide continuous latent representations of chemical space, targeted optimization within these representations often leads to degraded validity, loss of structural fidelity, or unstable behavior.
We introduce **MoltenFlow**, a modular framework that combines property-organized latent representations with flow-matching generative priors and gradient-based guidance. This formulation supports both conditioned generation and local optimization within a single latent-space framework. We show that guided latent flows enable efficient multi-objective molecular optimization under fixed oracle budgets with controllable trade-offs, while a learned flow prior improves unconditional generation quality.

Submission Track: Full Paper

Submission Category: AI-Guided Design

Submission Number: 21