新方法通过投机解码加速大语言模型推理 · 跟踪4个来源

报道来源 [4]

1. arXiv cs.CL TIER_1 English(EN) · Yu-Yang Qian, Hao-Cong Wu, Chen Chen, Jiacheng Sun, Zhenhua Dong, Peng Zhao, Zhi-Hua Zhou · 2026-07-22 04:00

   AdaFlash：通过 On-Policy 蒸馏扩散草稿器实现自适应推测解码

   arXiv:2607.19223v1 Announce Type: cross Abstract: Speculative decoding, in which a lightweight draft model first generates a draft sequence that is then verified in parallel by the target model, has become a prevalent paradigm for accelerating large language model inference. Rece…

2. arXiv cs.CL TIER_1 English(EN) · Zhibin Wang, Xuying Han, Zhaohua Yang, Fuliang Liu, Xue Li, Rong Gu, Sheng Zhong, Chen Tian · 2026-07-21 04:00

   SpecLA：面向线性注意力模型的 eficient 投机解码

   arXiv:2607.16673v1 Announce Type: new Abstract: Linear-attention models replace the growing KV cache with recurrent states, but autoregressive decoding still reads, updates, and writes these states one token at a time. Speculative decoding can reduce this cost by verifying severa…

3. arXiv cs.CL TIER_1 English(EN) · Miles Williams, Young D. Kwon, Rui Li, Alexandros Kouris, Stylianos I. Venieris · 2026-07-20 04:00

   具有推测性词汇的推测性解码

   arXiv:2602.13836v2 Announce Type: replace Abstract: Speculative decoding has rapidly emerged as a leading approach for accelerating language model (LM) inference, as it offers substantial speedups while yielding identical outputs. This relies upon a small draft model, tasked with…

4. Hugging Face Daily Papers TIER_1 English(EN) · 2026-07-12 09:10

   通过渐进式树草稿的推测解码解锁自回归语言模型的并行性

   Speculative decoding has significantly accelerated Large Language Model (LLM) inference by alleviating memory-bound bottlenecks. However, traditional speculative decoding typically relies on auxiliary draft modules, incurring significant training and communication overhead. Altho…