In the impending era of data deluge, escalating costs related to time and energy owing to data movement underscore the need for a departure from traditional systems. The pursuit of in-memory computing that emulates the high-density memory and energy-efficient information processing of the human brain is of paramount significance. At the forefront of this initiative are memtransistors, with an emphasis on those capable of processing multibit digital and analog data and offering the distinctive features of electrostatically tuning memory and learning behaviors at the device level. Herein, a conceptual overview of the materials and device architectures of memtransistors is presented to underscore their pivotal role in in-memory computing. The discussion includes the strategies for achieving their 3D integration and addresses pertinent challenges and opportunities from materials to chip level for advanced memory storage and neuromorphic computing applications.

中文翻译：

---

用于神经形态硬件和内存计算的记忆晶体管的兴起

在即将到来的数据洪流时代，数据移动导致的时间和能源成本不断上升，凸显了脱离传统系统的必要性。追求模拟人脑高密度内存和节能信息处理的内存计算具有至关重要的意义。该计划的最前沿是记忆晶体管，重点是那些能够处理多位数字和模拟数据并在设备级别提供静电调节记忆和学习行为的独特功能的记忆晶体管。本文对记忆晶体管的材料和器件架构进行了概念性概述，以强调它们在内存计算中的关键作用。讨论包括实现 3D 集成的策略，并解决从材料到芯片级的高级内存存储和神经形态计算应用的相关挑战和机遇。