---
id: 20261118-T0-10
title: "SpecRelay：边缘-云协同推测解码，省掉往返通信开销"
title_en: "SpecRelay Cuts Round-Trip Overhead in Edge-Cloud Speculative Decoding"
url: https://ai.daily.yangsir.net/daily/20261118-T0-10
issue_date: 2026-11-18
publish_date: 2026-11-17T05:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2611.18963
---

# SpecRelay：边缘-云协同推测解码，省掉往返通信开销

新论文提出 SpecRelay 方案，针对边缘-云推测解码的往返通信瓶颈。该架构将自回归草稿生成放在边缘设备，云端大模型并行验证草稿 token，两者在同步草稿-验证调度下重叠执行。此前方法受限于低延迟网络下的通信开销，SpecRelay 通过摊销或规避往返来提升整体效率。对端侧部署 LLM 的开发者来说，这意味着在保持云端验证质量的同时，减少边缘与云之间的等待时间。

## English Version

**SpecRelay Cuts Round-Trip Overhead in Edge-Cloud Speculative Decoding**

A new arXiv paper introduces SpecRelay, a scheme targeting the round-trip communication bottleneck in edge-cloud speculative decoding. The architecture runs autoregressive drafting on an edge device while a larger cloud model verifies draft tokens in parallel, overlapping the two under a synchronous draft-verify schedule. Prior methods were constrained by communication overhead under low-latency networks; SpecRelay amortizes or avoids the round trip to improve overall efficiency. For developers deploying LLMs on edge devices, this means less waiting between edge and cloud while keeping cloud-level verification quality.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2611.18963)

**详情页**：https://ai.daily.yangsir.net/daily/20261118-T0-10

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*