Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling

Published in ICLR, 2026

Recommended citation: Shengyin Sun, Yiming Li, Xing Li, Yingzhao Lian, Weizhe Lin, Huiling Zhen, Zhiyuan Yang, Xianzhi Yu, Chen Chen, Mingxuan Yuan, Chen Ma. (2026). Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling. International Conference on Learning Representations, 2026, 103030–103060. https://proceedings.iclr.cc/paper_files/paper/2026/hash/a8808b75b299d64a23255bc8d30fb786-Abstract-Conference.html