融合文本与图结构，提升分子GNN在材料科学中的预测能力
Understanding the Capabilities of Molecular Graph Neural Networks in Materials Science Through Multimodal Learning and Physical Context Encoding
- 用文本描述与分子图联合建模，补充几何信息
- 添加文本特征后电子性质预测准确率显著提升
- 不同模型表现相似，说明学习到的是共性表征
分子图神经网络（GNN）通常仅依赖基于XYZ的几何表示，忽略了公共数据库（如PubChem）中丰富的化学上下文信息。本文提出一种多模态框架，将IUPAC名称、分子式、理化性质、别名等文本描述与分子图结合。采用门控融合机制平衡几何与文本特征，使模型能够利用互补信息。在基准数据集上的实验表明，加入文本数据可显著提升部分电子性质的预测效果，而对其他性质改善有限。此外，不同GNN架构在相同目标上表现出相似的性能变化模式，说明它们学习的是类似表征而非不同的物理洞察。
原文摘要 · Abstract (English)
Molecular graph neural networks (GNNs) often focus exclusively on XYZ-based geometric representations and thus overlook valuable chemical context available in public databases like PubChem. This work introduces a multimodal framework that integrates textual descriptors, such as IUPAC names, molecular formulas, physicochemical properties, and synonyms, alongside molecular graphs. A gated fusion mechanism balances geometric and textual features, allowing models to exploit complementary information. Experiments on benchmark datasets indicate that adding textual data yields notable improvements for certain electronic properties, while gains remain limited for others. Furthermore, the GNN architectures display similar performance patterns (improving and deteriorating on analogous targets), suggesting they learn comparable representations rather than distinctly different physical insights.
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。