数据分布至关重要：通过协同利用正负样本数据增强数据驱动材料发现中的模型泛化能力
武一衡, 刘立军, 邓伟, 姜浩东, 李若彤, 胡铭, 彭博, 王大鹏
Data distribution matters: enhancing model generalization in datadriven materials discovery by intentionally exploiting both positive and negative data
Yiheng Wu, Lijun Liu, Wei Deng, Haodong Jiang, Ruotong Li, Ming Hu, Bo Peng, Dapeng Wang
物理化学学报
.
2026, (): 100291
.
DOI: 10.1016/j.actphy.2026.100291
