@article{saliencydrivendynamictokenpruningfor, title = {Saliency-driven Dynamic Token Pruning for Large Language Models}, author = {Yao Tao and Yehui Tang and Yun Wang and Mingjian Zhu and Hailin Hu and Yunhe Wang}, year = {2025}, eprint = {2504.04514}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2504.04514v2}, }