@article{efficientservingfordynamicagentworkflows, title = {Efficient Serving for Dynamic Agent Workflows with Prediction-based KV-Cache Management}, author = {Haoyu Zheng and Fangcheng Fu and Jia Wu and Binhang Yuan and Yongqiang Zhang and Hao Wang and Yuanyuan Zhu and Xiao Yan and Jiawei Jiang}, year = {2026}, eprint = {2605.06472}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2605.06472}, }