Papers › Improving Few- and Zero-Shot Reaction Template Prediction Using Modern Hopfield Networks

Improving Few- and Zero-Shot Reaction Template Prediction Using Modern Hopfield Networks

15 Jan 2022Journal of Chemical Information and Modeling 2022 1archive 2025-07-28

Philipp Seidl, Philipp Renz, Natalia Dyubankova, Paulo Neves, Jonas Verhoeven, Jörg K. Wegner, Marwin Segler, Sepp Hochreiter, and Günter Klambauer

Finding synthesis routes for molecules of interest is essential in the discovery of new drugs and materials. To find such routes, computer-assisted synthesis planning (CASP) methods are employed, which rely on a single-step model of chemical reactivity. In this study, we introduce a template-based single-step retrosynthesis model based on Modern Hopfield Networks, which learn an encoding of both molecules and reaction templates in order to predict the relevance of templates for a given molecule. The template representation allows generalization across different reactions and significantly improves the performance of template relevance prediction, especially for templates with few or zero training examples. With inference speed up to orders of magnitude faster than baseline methods, we improve or match the state-of-the-art performance for top-k exact match accuracy for k ≥ 3 in the retrosynthesis benchmark USPTO-50k. Code to reproduce the results is available at github.com/ml-jku/mhn-react.

PaperPDFCode

Code

ml-jku/mhn-react mentioned in paperpytorchNOASSERTION report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

RetrosynthesisSingle-step retrosynthesis

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Single-step retrosynthesis USPTO-50k MHNreact (reaction class unknown) Top-1 accuracy 51.8 #28 of 35 Archive leaderboard report
Single-step retrosynthesis USPTO-50k MHNreact (reaction class unknown) Top-10 accuracy 88.1 #28 of 35 Archive leaderboard report
Single-step retrosynthesis USPTO-50k MHNreact (reaction class unknown) Top-20 accuracy 92.0 #28 of 35 Archive leaderboard report
Single-step retrosynthesis USPTO-50k MHNreact (reaction class unknown) Top-3 accuracy 74.6 #28 of 35 Archive leaderboard report
Single-step retrosynthesis USPTO-50k MHNreact (reaction class unknown) Top-5 accuracy 81.2 #28 of 35 Archive leaderboard report
Single-step retrosynthesis USPTO-50k MHNreact (reaction class unknown) Top-50 accuracy 94.0 #28 of 35 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections