Paper recorded by Signals 4 on 2026-10-14 in cs.CL. Abstract reproduced from arXiv; link to the original below.
Published 2026-10-14 on arXiv · recorded by Signals 4 on 2026-10-15
Category: cs.CL · 自然语言处理 · first seen 2026-10-15
Speculative decoding accelerates LLM generation through lightweight drafting and parallel verification. It is particularly promising for long inputs, where KV-cache access costs can be amortized over multiple accepted tokens. Yet most drafters are trained on short sequences and degrade on long inputs, while long-context training is costly. Existing training-free approaches restrict the context win