@article{headawarekvcachecompressionforefficient, title = {Head-Aware KV Cache Compression for Efficient Visual Autoregressive Modeling}, author = {Ziran Qin and Youru Lv and Mingbao Lin and Zeren Zhang and Danping Zou and Weiyao Lin}, year = {2025}, eprint = {2504.09261}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2504.09261v1}, }