<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="metel-etal-2024-draft">
    <titleInfo>
        <title>Draft on the Fly: Adaptive Self-Speculative Decoding using Cosine Similarity</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Michael</namePart>
        <namePart type="given">R</namePart>
        <namePart type="family">Metel</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Peng</namePart>
        <namePart type="family">Lu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Boxing</namePart>
        <namePart type="family">Chen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Mehdi</namePart>
        <namePart type="family">Rezagholizadeh</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ivan</namePart>
        <namePart type="family">Kobyzev</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2024</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yaser</namePart>
            <namePart type="family">Al-Onaizan</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mohit</namePart>
            <namePart type="family">Bansal</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yun-Nung</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Miami, Florida, USA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>We present a simple on the fly method for faster inference of large language models. Unlike other (self-)speculative decoding techniques, our method does not require fine-tuning or black-box optimization to generate a fixed draft model, relying instead on simple rules to generate varying draft models adapted to the input context. We show empirically that our light-weight algorithm is competitive with the current SOTA for self-speculative decoding, while being a truly plug-and-play method.</abstract>
    <identifier type="citekey">metel-etal-2024-draft</identifier>
    <identifier type="doi">10.18653/v1/2024.findings-emnlp.124</identifier>
    <location>
        <url>https://aclanthology.org/2024.findings-emnlp.124</url>
    </location>
    <part>
        <date>2024-11</date>
        <extent unit="page">
            <start>2267</start>
            <end>2272</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Draft on the Fly: Adaptive Self-Speculative Decoding using Cosine Similarity
%A Metel, Michael R.
%A Lu, Peng
%A Chen, Boxing
%A Rezagholizadeh, Mehdi
%A Kobyzev, Ivan
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Findings of the Association for Computational Linguistics: EMNLP 2024
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F metel-etal-2024-draft
%X We present a simple on the fly method for faster inference of large language models. Unlike other (self-)speculative decoding techniques, our method does not require fine-tuning or black-box optimization to generate a fixed draft model, relying instead on simple rules to generate varying draft models adapted to the input context. We show empirically that our light-weight algorithm is competitive with the current SOTA for self-speculative decoding, while being a truly plug-and-play method.
%R 10.18653/v1/2024.findings-emnlp.124
%U https://aclanthology.org/2024.findings-emnlp.124
%U https://doi.org/10.18653/v1/2024.findings-emnlp.124
%P 2267-2272

Markdown (Informal)

[Draft on the Fly: Adaptive Self-Speculative Decoding using Cosine Similarity](https://aclanthology.org/2024.findings-emnlp.124) (Metel et al., Findings 2024)

• Draft on the Fly: Adaptive Self-Speculative Decoding using Cosine Similarity (Metel et al., Findings 2024)

ACL

• Michael R. Metel, Peng Lu, Boxing Chen, Mehdi Rezagholizadeh, and Ivan Kobyzev. 2024. Draft on the Fly: Adaptive Self-Speculative Decoding using Cosine Similarity. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 2267–2272, Miami, Florida, USA. Association for Computational Linguistics.