atlas

Research map / Graph learning / Graph algorithms and clustering

Predicting Organic Reaction Outcomes with Weisfeiler-Lehman Network

NeurIPS 2017 · Graph learning › Graph algorithms and clustering · approach: Neural combinatorial optimization on graphs · 218 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add global attention context to the reaction-center identifier (ablation)WLN reaction-center model · USPTO-15K, K=8coverage: 86.1 -> 90.1Helped
Add global attention context to the reaction-center identifier (ablation)WLN reaction-center model · USPTO, K=8coverage: 87.2 -> 92.0Helped
Use a WLDN difference graph rather than sum-pooling difference vectors (ablation)Candidate ranking on USPTO-15K · USPTO-15KP@1: 74.9 -> 76.7Helped
Add the true product when reaction-center candidate generation misses it (ablation)WLDN candidate ranking · USPTO-15KP@1: 76.7 -> 84.1Helped
Add the true product when reaction-center candidate generation misses it (ablation)WLDN candidate ranking · USPTOP@1: 79.6 -> 83.9Helped
Replace template generation and matching with reaction-center prediction and candidate enumerationTemplate-free reaction prediction · USPTO candidate generationspeed: 140 times fasterHelped
Use the template-free model instead of the average chemist predictionReaction outcome prediction · USPTO test set, 80 reactionsaccuracy: 48.2%; 69.1%Helped

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy