X-CoSD: Communication-Efficient Cross-Vocabulary Collaborative Speculative Decoding

Excerpt

arXiv:2609.09166v1 Announce Type: new Abstract: This paper investigates collaborative speculative decoding (CoSD), a distributed large language model (LLM) inference framework in which an on-device small language model (SLM) drafts candidate tokens and a server LLM verifies them. Existing CoSD methods assume a shared vocabulary between the SLM and the LLM and incur substantial communication load because residual resampling requires token distribution exchange between the user device and the edge