3 papers
cs.AR2025
Hardware Efficient Accelerator for Spiking Transformer With Reconfigurable Parallel Time Step Computing
Bo-Yu Chen, Tian-Sheuan Chang
This paper introduces the first low-power hardware accelerator for Spiking Transformers, an emerging alternative to traditional artificial neural networks. By modifying the base Sp…
cs.LG2025
An Efficient Data Reuse with Tile-Based Adaptive Stationary for Transformer Accelerators
Tseng-Jen Li, Tian-Sheuan Chang
Transformer-based models have become the \textit{de facto} backbone across many fields, such as computer vision and natural language processing. However, as these models scale in s…
cs.AR2025
A Low-Power Sparse Deep Learning Accelerator with Optimized Data Reuse
Kai-Chieh Hsu, Tian-Sheuan Chang
Sparse deep learning has reduced computation significantly, but its irregular non-zero data distribution complicates the data flow and hinders data reuse, increasing on-chip SRAM a…