提出新型注意力机制，提升小样本分子属性预测精度
Cardinality-Preserving Attention Channels for Graph Transformers in Molecular Property Prediction
- 设计动态保留支持集大小信号的注意力通道
- 11个基准上均优于基线模型，小样本下效果更显著
- 适合药物发现中的分子性质预测任务
分子属性预测在标签数据稀缺的药物发现中至关重要。本文提出CardinalGraphFormer，一种引入查询相关性保持基数注意力（CPA）通道的图变换器，可保留与静态中心性嵌入互补的动态支持集信号。方法结合结构化稀疏注意力与Graphormer启发的偏置（最短路径距离、中心性、直接键特征），并采用统一双目标自监督预训练（掩码重建与增强视图对比对齐）。在涵盖MoleculeNet、OGB和TDC ADMET的11个公开基准上评估，经一致预训练、优化和超参数调优后，性能持续优于匹配基线。严谨消融实验确认了CPA的有效性，并排除简单大小捷径可能。代码与可复现资源已提供。
原文摘要 · Abstract (English)
Molecular property prediction is crucial for drug discovery when labeled data are scarce. This work presents CardinalGraphFormer, a graph transformer augmented with a query-conditioned cardinality-preserving attention (CPA) channel that retains dynamic support-size signals complementary to static centrality embeddings. The approach combines structured sparse attention with Graphormer-inspired biases (shortest-path distance, centrality, direct-bond features) and unified dual-objective self-supervised pretraining (masked reconstruction and contrastive alignment of augmented views). Evaluation on 11 public benchmarks spanning MoleculeNet, OGB, and TDC ADMET demonstrates consistent improvements over protocol-matched baselines under matched pretraining, optimization, and hyperparameter tuning. Rigorous ablations confirm CPA's contributions and rule out simple size shortcuts. Code and reproducibility artifacts are provided.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。