With the rapid development of machine learning, the demand for high-efficient computing becomes more and more urgent. To break the bottleneck of the traditional Von Neumann architecture, computing-in-memory (CIM) has attracted increasing attention in recent years. In this work, to provide a feasible CIM solution for the large-scale neural networks (NN) requiring continuous weight updating in online training, a flash-based computing-in-memory with high endurance (10⁹ cycles) and ultra-fast programming speed is investigated. On the one hand, the proposed programming scheme of channel hot electron injection (CHEI) and hot hole injection (HHI) demonstrate high linearity, symmetric potentiation, and a depression process, which help to improve the training speed and accuracy. On the other hand, the low-damage programming scheme and memory window (MW) optimizations can suppress cell degradation effectively with improved computing accuracy. Even after 10⁹ cycles, the leakage current (I_off) of cells remains sub-10pA, ensuring the large-scale computing ability of memory. Further characterizations are done on read disturb to demonstrate its robust reliabilities. By processing CIFAR-10 tasks, it is evident that ~90% accuracy can be achieved after 10⁹ cycles in both ResNet50 and VGG16 NN. Our results suggest that flash-based CIM has great potential to overcome the limitations of traditional Von Neumann architectures and enable high-performance NN online training, which pave the way for further development of artificial intelligence (AI) accelerators.

Mid-spatial frequency wavefront deformation can be deleterious for the operation of high-energy laser systems. When fluid cooled high-repetition-rate amplifiers are used, the coolant flow is likely to induce such detrimental mid-spatial frequency wavefront deformations. Here, we describe the design and performance of a 90 mm × 90 mm aperture, liquid-cooled Nd:phosphate split-slab laser amplifier pumped by flash-lamps. The performance of the system is evaluated in terms of wavefront aberration and gain at repetition rates down to 1 shot per minute. The results show that this single cooled split-slab system exhibits low wavefront distortions in the medium to large period range, compatible with a focus on target, and despite the use of liquid coolant traversed by both pump and amplified wavelengths. This makes it a potential candidate for applications in large high-energy laser facilities.

报道了基于常温射频直线加速器建成的10 MeV光子FLASH放疗射线源样机，采用高平均电流电子束轰击高速旋转辐射转化靶，在距离靶点1 m远处的固体水模内，X射线剂量率达到80.5 Gy/s，达到未来临床实验与推广所需剂量率阈值。

为了系统研究FLASH效应的放射生物学机制，需要一个可以进行超高剂量率细胞辐照实验的平台，该实验平台应具有稳定、大小合适且大范围可调的剂量和平均剂量率。基于国产7 MeV医用质子直线注入器，使用蒙特卡罗程序FLUKA设计并优化了一个单散射照射头。该照射头的材料为40 μm厚度的钽箔，它既充当真空窗又充当散射体，源皮距为26 cm。经过模拟验证，该平台可提供直径为2 cm的照射野，剂量均匀度为4.9%。通过调整单质子脉冲的流强（0.1~1 mA）和脉宽（20~200 μs），该实验平台的平均剂量和平均剂量率可以在6~667 Gy和3.3×10⁵~3.3×10⁶ Gy·s^-1之间调节。基于此，设计了使用单脉冲穿透模式辐照单层细胞的实验，剂量率为3.3×10⁵ Gy·s^-1，剂量在7~40 Gy范围内变化。该实验平台可以探究细胞FLASH效应的总剂量依赖关系，为揭示FLASH效应的机制提供支持。

陶瓷材料因具有良好的机械性能、抗腐蚀性、耐高温性及抗氧化性等, 被广泛应用于航空航天、医疗、能源交通等领域, 陶瓷材料自身及其与金属材料的连接技术对于实际工程应用具有重要意义。由于部分陶瓷材料与电场的特殊作用机理, 将外加电场应用于陶瓷材料的连接技术中, 可以获得多种普通连接技术所不具备的优势, 如连接温度较低和连接时间较短等, 这就催生了新型陶瓷材料电场辅助连接技术。本文着重梳理了陶瓷及陶瓷基复合材料电场辅助连接技术的研究现状, 对近年来电场辅助连接技术的研究进展进行了综述, 重点介绍了电场辅助扩散连接(Electric field-assisted diffusion bonding, FDB) 技术、放电等离子体烧结 (Spark plasma sintering, SPS)连接技术以及新型低温快速闪连接(Flash joining, FJ) 技术的连接机理、典型界面微观结构、接头强度及影响因素等, 阐述了不同电场辅助连接技术的适用范围和局限性, 并对陶瓷材料电场辅助连接技术的发展进行了展望。

近年来，一项被称为FLASH放疗的新技术具有将辐射的治疗增益比推向一个新高度的潜力。FLASH放疗是一种以超高剂量率照射为主要特征的放疗技术，能显著减轻辐射对正常组织的损伤，同时保留辐射对肿瘤的杀伤能力。一方面，作为一种新兴的放疗技术，FLASH放疗的生物学机制尚未被阐明，这阻碍了其临床转化；另一方面，超高剂量率照射技术的临床应用也存在许多技术困难。本综述通过归纳FLASH放射生物学机制的研究进展，总结FLASH放疗治疗计划所面临的困难及可能的解决方案，旨在为后续FLASH放疗的临床转化提供参考。