DiffER$^2$: Diffusion Ensembles for Retrosynthesis Prediction with SMILES Adapted Particle Guidance
Its authors withdrew this paper, so there is no full text to read or summarize.
Lists
Only you see this
Abstract
Computational retrosynthesis prediction has the potential to reduce development time for newly discovered drugs by automatically generating potential reactions for a target product. Data-driven approaches commonly treat this as a sequence-to-sequence generation task on SMILES strings. In this work, we construct ensembles of discrete-time and continuous-time diffusion models for molecular generation and incorporate guidance mechanisms for improved output diversity. We propose an adaptation of particle guidance to SMILES sequence generation which significantly improves the number of unique molecules generated by diffusion ensembles while increasing top-k accuracy. These results further expand the efficacy of discrete diffusion for SMILES generation, and our empirical analyses offer new insights into the capabilities of diffusion models for chemical retrosynthesis.
BibTeX
@misc{current2026differ,
author = {Sean Current and Daniel Adu-Ampratwum and Xia Ning and srinivasan parthasarathy},
title = {{DiffER$^2$: Diffusion Ensembles for Retrosynthesis Prediction with SMILES Adapted Particle Guidance}},
year = {2026},
url = {https://openreview.net/forum?id=Nh1mW8sYd3},
howpublished = {OpenReview},
note = {Submitted to ICLR 2026},
}