作者位次 #5/6ACL 2026Poster· Natural Language Processing
EDSD : Entropy-Driven Design for Faster Speculative Decoding
Longkai Cheng · Ximing Wang · Jiangcai Zhu · Kailai Shao · Chao Chen · Haixiang Hu
摘要
Speculative decoding has emerged as a promising paradigm for accelerating large language model inference by leveraging a lightweight draft model to generate multiple candidate tokens. However, existing methods often incur substantial training overhead to mitigate information misalignment between...