作者位次 #5/10ICML 2026Poster· Deep Learning / Large Language Models
FAFO: Lossy KV Cache Compression for Lossless Inference Acceleration via Draftless Fumble Decoding
Hoang Anh Duy Le · Shaochen (Henry) Zhong · Yifan Lu · Yingtong Dou · Jiayi Yuan · Yu-Neng Chuang · Xiran Fan · Guanchu Wang · Yuzhong Chen · Xia Hu