Abstract
Graph self-supervised learning (GSSL) has demonstrated strong potential for generating expressive graph embeddings without the need for human annotations, making it particularly valuable in domains with high labeling costs such as molecular graph analysis. However, existing GSSL methods mostly focus on node- or edge-level information, often ignoring chemically relevant substructures which strongly influence molecular properties. In this work, we propose Graph Semantic Predictive Network (GraSPNet), a hierarchical self-supervised framework that explicitly models both atomic-level and fragment-level semantics. GraSPNet decomposes molecular graphs into chemically meaningful fragments without predefined vocabularies and learns node- and fragment-level representations through multi-level message passing with masked semantic prediction at both levels. This hierarchical semantic supervision enables GraSPNet to learn multi-resolution structural information that is both expressive and transferable. Extensive experiments on multiple molecular property prediction benchmarks demonstrate that GraSPNet learns chemically meaningful representations and consistently outperforms state-of-the-art GSSL methods in transfer learning settings.
Abstract (translated)
图自监督学习（GSSL）在生成无需人工标注的富有表现力的图嵌入方面表现出巨大潜力，特别是在分子图分析等标签成本高昂的应用领域中尤其有价值。然而，现有的大多数GSSL方法主要关注节点级或边级信息，往往忽略了强烈影响分子属性的化学相关子结构。在此研究中，我们提出了Graph Semantic Predictive Network（GraSPNet），这是一种层次化的自监督框架，它显式地对原子级和片段级语义进行建模。GraSPNet将分子图分解为具有化学意义的片段，并且不依赖预定义词汇表，在多层级信息传递过程中通过在节点级别和片段级别的掩码语义预测学习节点表示和片段表示。这种层次化的语义监督使得GraSPNet能够从数据中学习到兼具表达性和可迁移性的多层次结构信息。 在多个分子属性预测基准测试中的广泛实验表明，GraSPNet学到了具有化学意义的表征，并且在迁移学习设置下始终优于最先进的GSSL方法。
URL
https://arxiv.org/abs/2602.20344
