@article{solehardwaresoftwarecodesignofsoftmaxand, title = {SOLE: Hardware-Software Co-design of Softmax and LayerNorm for Efficient Transformer Inference}, author = {Wenxun Wang and Shuchang Zhou and Wenyu Sun and Peiqin Sun and Yongpan Liu}, year = {2025}, eprint = {2510.17189}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2510.17189}, }