On January 8, Professor Wang Xinran's collaborative team from Nanjing University published groundbreaking research in the prestigious international journal Nature Electronics. Titled "An Index-Free Sparse Neural Network Using Two-Dimensional Semiconductor Ferroelectric Field-Effect Transistors," this study introduces an innovative hardware solution for sparse neural networks based on two-dimensional materials. This approach addresses the long-standing issue of mismatch between sparse neural network software and hardware. The team pioneered the "in-memory sparse" computing architecture, leveraging molybdenum disulfide ferroelectric transistor technology to achieve index-free sparse neural network training, significantly reducing training overhead. Furthermore, they developed a co-optimization method for hardware and software, enhancing system performance even further. This breakthrough is poised to offer fresh perspectives and technical advancements for the development of artificial intelligence hardware.