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

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 insufficient structural context modeling & 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 Relational and Compact context Graph, named RECOG, to comprehensively exploit the context graph for expressive molecular property prediction. Specifically, the proposed RECOG contains two core modules: a (1) crossproperty relational learning module to better model the structural and relational context information, and a (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. Extensive experiments on multiple datasets demonstrate that RECOG consistently outperforms state-of-the-art methods, validating its superiority. Code is available at the repository.