Interpretable and Steerable Molecular Graph Generation using Concept Bottlenecks

Abstract:Deep graph generators (DGGs) largely lack interpretability, which is a significant limitation for one of their most crucial applications, drug discovery, where interpretable and transparent generation could provide greater control over the desirability of generated molecules, thereby reducing trial-and-error in candidate selection. Common approaches to interpretability, such as disentanglement, cannot guarantee that the learned representations are human-interpretable. To address this, we propose incorporating concept bottlenecks into DGGs, allowing users to interpret, debug, and steer the generation process. We show that this consistently improves distributional fidelity (lower FCD), with only modest, non-systematic effects on validity and uniqueness. We further explore Concept Controller, an optimization-based steering approach that has proven to be effective for steerable image generation. We evaluate our approach on a one-shot molecular graph generation framework, though the underlying ideas can be transferred to sequential graph generation as well.

Accepted at AI for Drug Discovery Workshop @ NeurIPS 2026. Additional experiments added

Machine Learning (cs.LG)

arXiv:2610.34227 [cs.LG]

(or arXiv:2610.34227v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2610.34227 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.