@inproceedings{chen-etal-2026-cachespec, title = "{C}ache{S}pec: Finding the Sweet Spot for Small Models in Large Language Models", author = "Chen, Jingquan and Feng, Jie and Piao, Jinghua and Hu, Shaogang and Li, Yong", editor = "Lee, Hung-Yi and Dev, Sunipa and Li, Jessy and Elliott, Desmond", booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing", month = oct, year = "2026", address = "Budapest, Hungary", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2026.emnlp-main.1223/", pages = "27911--27928", ISBN = "979-8-89176-459-0" }