HiRes: Inspectable Precedent Memory for Reaction Condition Recommendation
For chemists and synthesis planning, HiRes offers accurate condition recommendations with interpretable chemical precedents, bridging the gap between predictive performance and practical usability.
HiRes is a retrieval-augmented reaction condition recommendation system that achieves state-of-the-art performance on USPTO-Condition, with Catalyst, Solvent, and Reagent top-1 accuracies of 0.929, 0.534, and 0.530 respectively, and provides inspectable precedents for predictions.
Reaction condition recommendation sits immediately after retrosynthetic disconnection selection, and in practice, chemists require both accurate predictions and the precedents that justify them. We present HiRes (Hierarchical Reaction Representations), a retrieval-augmented condition recommendation system whose learned reaction space serves as both a classifier feature and an inspectable precedent memory. The model combines a graph encoder, transformation-aware cross-attention, multi-stream reaction fusion, and a k-NN retrieval layer. HiRes achieves state-of-the-art performance among primary-slot USPTO-Condition models, reaching Catalyst, Solvent, and Reagent top-1 accuracies (Acc@1) of 0.929, 0.534, and 0.530 respectively. It ties the best reported baseline on Catalyst while outperforming models such as REACON on Solvent and Reagent. Furthermore, paired bootstrap analysis demonstrates that integrating retrieval with learned condition heads provides statistically significant gains for solvent and reagent selection over purely parametric approaches. Ultimately, HiRes bridges the gap between predictive accuracy and chemical interpretability, offering a single representation that supplies both competitive recommendations and the concrete chemical precedents necessary for practical synthesis planning.