为解决传统冯·诺依曼架构的能效瓶颈，研究人员开发了一种CMOS兼容的闪存门控晶闸管神经形态模块（FGTNM），集成量化、非线性激活和最大池化功能于单一模块。该模块面积仅53平方微米，能耗低至9.1飞焦/次，在CIFAR-10分类任务中实现89.97%的准确率，为存内计算（IMC）硬件设计提供了突破性方案。

A pioneering review published in Nano-Micro Letters provides a thorough overview of the significance of two-dimensional (2D) materials in defining the future of electronics and integrated circuits ...

This meta-analysis from two nationwide cohorts indicates that SSRI and SNRI augmentation therapies for people with ...

A new study published in JAMA Network Open suggests that brain imaging could help identify who is most likely to benefit from ...