Local-Global Multimodal Contrastive Learning for Molecular Property Prediction
局部-全局多模态对比学习用于分子性质预测
机构 * School of Mathematical Sciences(数学科学学院) ; University of Electronic Science and Technology of China(电子科技大学) ; Department of Computer Science and Engineer(计算机科学与工程系) ; Oakland University(奥克兰大学) ; Department of Electrical and Computer Engineering(电气与计算机工程系) ; College of Management Science(管理科学学院) ; Chengdu University of Technology(成都理工大学)
专题命中 融合架构与评测 :multimodal fusion(abstract)
AI总结 LGM-CL通过局部-全局多模态对比学习框架,整合分子结构和化学语义信息,提升分子性质预测的准确性与性能。
Comments 16 pages, 9 figures. Submitted to Briefings in Bioinformatics