RetroGFN: Diverse and Feasible Retrosynthesis using GFlowNets

Piotr Gainski, Michal Koziarski, Krzysztof Maziarz, Marwin H. S. Segler, Jacek Tabor, Marek Smieja

Abstract

Single-step retrosynthesis aims to predict a set of reactions that lead to the creation of a target molecule, which is a crucial task in molecular discovery. Although a target molecule can often be synthesized with multiple different reactions, it is not clear how to verify the feasibility of a reaction, because the available datasets cover only a tiny fraction of the possible solutions. Consequently, the existing models are not encouraged to explore the space of possible reactions sufficiently. In this paper, we propose a novel single-step retrosynthesis model, RetroGFN, that can explore outside the limited dataset and return a diverse set of feasible reactions by leveraging a feasibility proxy model during the training. We show that RetroGFN achieves competitive results on standard top-k accuracy while outperforming existing methods on round-trip accuracy. Moreover, we provide empirical arguments in favor of using round-trip accuracy which expands the notion of feasibility with respect to the standard top-k accuracy metric.
BibTeX
@article{gainski2025retrogfn,
  author = {Piotr Gainski and Michal Koziarski and Krzysztof Maziarz and Marwin H. S. Segler and Jacek Tabor and Marek Smieja},
  title = {{RetroGFN: Diverse and Feasible Retrosynthesis using GFlowNets}},
  journal = {Information Sciences},
  year = {2025},
  doi = {10.1016/j.ins.2025.122194},
}
Figure