Matzinger, H., & Komatuzaki, A. (2026). Lower bound on the information-loss incurred by compressing a depth-2 feed-forward layer of transformer into a single fully connected layer. Journal of Algorithms & Computational Technology, 20, 1. https://doi.org/10.1177/17483026261455599
Chicago Style (17th ed.) CitationMatzinger, Heinrich, and Aran Komatuzaki. "Lower Bound on the Information-loss Incurred by Compressing a Depth-2 Feed-forward Layer of Transformer into a Single Fully Connected Layer." Journal of Algorithms & Computational Technology 20 (2026): 1. https://doi.org/10.1177/17483026261455599.
MLA (9th ed.) CitationMatzinger, Heinrich, and Aran Komatuzaki. "Lower Bound on the Information-loss Incurred by Compressing a Depth-2 Feed-forward Layer of Transformer into a Single Fully Connected Layer." Journal of Algorithms & Computational Technology, vol. 20, 2026, p. 1, https://doi.org/10.1177/17483026261455599.