ITA: An Energy-Efficient Attention and Softmax Accelerator for Quantized Transformers

Gamze Islamoglu, Moritz Scherer, Gianna Paulin, Tim Fischer 0001, Victor J. B. Jung, Angelo Garofalo, Luca Benini. ITA: An Energy-Efficient Attention and Softmax Accelerator for Quantized Transformers. In IEEE/ACM International Symposium on Low Power Electronics and Design, ISLPED 2023, Vienna, Austria, August 7-8, 2023. pages 1-6, IEEE, 2023. [doi]

Abstract

Abstract is missing.