ZHOU, Xichuan; LIU, Kui; SHI, Cong; LIU, Haijun; LIU, Ji. Optimizing Information Theory Based Bitwise Bottlenecks for Efficient Mixed-Precision Activation Quantization. Proceedings of the AAAI Conference on Artificial Intelligence, [S. l.], v. 35, n. 4, p. 3590–3598, 2021. DOI: 10.1609/aaai.v35i4.16474. Disponível em: https://ojs.aaai.org/index.php/AAAI/article/view/16474. Acesso em: 13 may. 2026.