Huang, Xinhao, et al. “SoLA: Leveraging Soft Activation Sparsity and Low-Rank Decomposition for Large Language Model Compression”. Proceedings of the AAAI Conference on Artificial Intelligence, vol. 39, no. 16, Apr. 2025, pp. 17494-02, doi:10.1609/aaai.v39i16.33923.