@article{isyournpureadyforllmsdissectingthehidden, title = {Is Your NPU Ready for LLMs? Dissecting the Hidden Efficiency Bottlenecks in Mobile LLM Inference}, author = {Guanyu Cai and Ruiming Tian and Lang Yang and Zhouhong Ren and Jinliang Yuan and Lingkun Li and Jiliang Wang}, year = {2026}, eprint = {2607.05475}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2607.05475}, }