ReCoG: Relational and Compact Context Graph Learning for Few-shot Molecular Property Prediction

2026-05-13

AI Summary

This work addresses the challenges of insufficient structural context modeling and redundant auxiliary information in few-shot molecular property prediction by proposing a novel paradigm that jointly models relational structure and compresses task-irrelevant signals. The approach constructs a context graph that is both relationally expressive and compact, integrating cross-property relational learning with an information bottleneck mechanism to enhance molecular representation quality. Theoretical analysis and empirical evaluations demonstrate that this method significantly improves prediction accuracy and generalization capability in few-shot settings.

Abstract

Few-shot molecular property prediction (FSMPP) is essential in drug discovery and materials design, where high-quality labeled data are often scarce and expensive to obtain. Despite the promising performance of existing methods, especially context-aware methods, they still face two-fold severe challenges with \textit{insufficient structural context modeling} \& \textit{redundant auxiliary context learning}, leading to inadequate context graph exploration and ineffective information utilization for effective molecule representation learning. To address these, in this paper, we propose a novel framework by learning on \textbf{\underline{Re}}lational and \textbf{\underline{C}}ompact c\textbf{\underline{o}}ntext \textbf{\underline{G}}raph, named \textbf{\method}, to comprehensively exploit the context graph for expressive molecular property prediction. Specifically, the proposed \method contains two core modules: a \textbf{(1) cross-property relational learning module} to better model the structural and relational context information, and a \textbf{(2) context graph information bottleneck module} to adaptively suppress irrelevant auxiliary signals for compact context information utilization, followed by a detailed theoretical demonstration regarding the importance of joint relational and compact knowledge extraction in context graphs.

Problem

• few-shot molecular property prediction
• structural context modeling
• redundant auxiliary context
• context graph
• molecular representation learning

Innovation

• few-shot learning
• molecular property prediction
• context graph
• relational learning
• information bottleneck