Template-based molecular generation offers a promising avenue for drug design by ensuring generated compounds are synthetically accessible through predefined reaction templates and building blocks. In this work, we tackle three core challenges in template-based GFlowNets: (1) minimizing synthesis cost, (2) scaling to large building block libraries, and (3) effectively utilizing small fragment sets. We propose Recursive Cost Guidance, a backward policy framework that employs auxiliary machine learning models to approximate synthesis cost and viability. This guidance steers generation toward low-cost synthesis pathways, significantly enhancing cost-efficiency, molecular diversity, and quality, especially when paired with an Exploitation Penalty that balances the trade-off between exploration and exploitation. To enhance performance in smaller building block libraries, we develop a Dynamic Library mechanism that reuses intermediate high-reward states to construct full synthesis trees. Our approach establishes state-of-the-art results in template-based molecular generation.
@article{arxiv.2506.19865,
title = {Scalable and Cost-Efficient de Novo Template-Based Molecular Generation},
author = {Piotr Gaiński and Oussama Boussif and Andrei Rekesh and Dmytro Shevchuk and Ali Parviz and Mike Tyers and Robert A. Batey and Michał Koziarski},
journal= {arXiv preprint arXiv:2506.19865},
year = {2025}
}