Abstract: This paper presents PALS, a high-speed and resourceefficient ML-DSA accelerator with a fully pipelined architecture, supporting all security levels. Implementation on Artix-7 and UltraScale+ ...