@inproceedings{liu-etal-2026-logitspec,
    title = "{L}ogit{S}pec: Accelerating Retrieval-based Speculative Decoding via Next Next Token Speculation",
    author = "Liu, Tianyu  and
      Lv, Qitan  and
      Li, Hao  and
      Gao, Xing  and
      Sun, Xiao  and
      Sun, Xiaoyan",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.1655/",
    pages = "33070--33092",
    ISBN = "979-8-89176-395-1",
    abstract = "Speculative decoding (SD), where a small draft model is employed to propose *draft* tokens in advance and then the target model validates them in parallel, has emerged as a promising technique for LLM inference acceleration. Many endeavors to improve SD are to eliminate the need for a draft model and generate draft tokens in a retrieval-based manner in order to further alleviate the drafting overhead and significantly reduce the difficulty in deployment and applications. However, retrieval-based SD relies on a matching paradigm to retrieve the most relevant reference as the draft tokens, where these methods often fail to find matched and accurate draft tokens. To address this challenge, we propose *LogitSpec* to effectively expand the retrieval range and find the most relevant reference as drafts. *LogitSpec* is motivated by the observation that the logit of the last token can not only predict **the next token**, but also speculate **the next next token**. Specifically, *LogitSpec* generates draft tokens in two steps: (1) utilizing the last logit to speculate the next next token; (2) retrieving relevant reference for both the next token and the next next token. *LogitSpec* is training-free and plug-and-play, which can be easily integrated into existing LLM inference frameworks. Extensive experiments on a wide range of text generation benchmarks demonstrate that *LogitSpec* can achieve up to 2.61{\texttimes} speedup and 3.28 mean accepted tokens per decoding step."
}