Paper ID: 2409.10489

Flash STU: Fast Spectral Transform Units

Y. Isabel Liu, Windsor Nguyen, Yagiz Devre, Evan Dogariu, Anirudha Majumdar, Elad Hazan

This paper describes an efficient, open source PyTorch implementation of the Spectral Transform Unit. We investigate sequence prediction tasks over several modalities including language, robotics, and simulated dynamical systems. We find that for the same parameter count, the STU and its variants outperform the Transformer as well as other leading state space models across various modalities.

Submitted: Sep 16, 2024