<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zuhri-etal-2025-mlkv">
    <titleInfo>
        <title>MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer Decoding</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Zayd</namePart>
        <namePart type="given">Muhammad</namePart>
        <namePart type="given">Kawakibi</namePart>
        <namePart type="family">Zuhri</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Muhammad</namePart>
        <namePart type="given">Farid</namePart>
        <namePart type="family">Adilazuarda</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ayu</namePart>
        <namePart type="family">Purwarianti</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Alham</namePart>
        <namePart type="given">Fikri</namePart>
        <namePart type="family">Aji</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-04</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: NAACL 2025</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Luis</namePart>
            <namePart type="family">Chiruzzo</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Alan</namePart>
            <namePart type="family">Ritter</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Lu</namePart>
            <namePart type="family">Wang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Albuquerque, New Mexico</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-195-7</identifier>
    </relatedItem>
    <abstract>Auto-regressive inference of transformers benefit greatly from Key-Value (KV) caching, but can lead to major memory bottlenecks as model size, batch size, and sequence length grow at scale. We introduce Multi-Layer Key-Value (MLKV) sharing, a novel approach extending KV sharing across transformer layers to reduce memory usage beyond what was possible with Multi-Query Attention (MQA) and Grouped-Query Attention (GQA). Evaluations on various NLP benchmarks and inference metrics using uptrained Pythia-160M variants demonstrate that MLKV significantly reduces memory usage with minimal performance loss, reducing KV cache size down to a factor of 6x compared to MQA. These results highlight MLKV’s potential for efficient deployment of transformer models at scale.</abstract>
    <identifier type="citekey">zuhri-etal-2025-mlkv</identifier>
    <identifier type="doi">10.18653/v1/2025.findings-naacl.305</identifier>
    <location>
        <url>https://aclanthology.org/2025.findings-naacl.305/</url>
    </location>
    <part>
        <date>2025-04</date>
        <extent unit="page">
            <start>5516</start>
            <end>5525</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer Decoding
%A Zuhri, Zayd Muhammad Kawakibi
%A Adilazuarda, Muhammad Farid
%A Purwarianti, Ayu
%A Aji, Alham Fikri
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Findings of the Association for Computational Linguistics: NAACL 2025
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-195-7
%F zuhri-etal-2025-mlkv
%X Auto-regressive inference of transformers benefit greatly from Key-Value (KV) caching, but can lead to major memory bottlenecks as model size, batch size, and sequence length grow at scale. We introduce Multi-Layer Key-Value (MLKV) sharing, a novel approach extending KV sharing across transformer layers to reduce memory usage beyond what was possible with Multi-Query Attention (MQA) and Grouped-Query Attention (GQA). Evaluations on various NLP benchmarks and inference metrics using uptrained Pythia-160M variants demonstrate that MLKV significantly reduces memory usage with minimal performance loss, reducing KV cache size down to a factor of 6x compared to MQA. These results highlight MLKV’s potential for efficient deployment of transformer models at scale.
%R 10.18653/v1/2025.findings-naacl.305
%U https://aclanthology.org/2025.findings-naacl.305/
%U https://doi.org/10.18653/v1/2025.findings-naacl.305
%P 5516-5525

Markdown (Informal)

[MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer Decoding](https://aclanthology.org/2025.findings-naacl.305/) (Zuhri et al., Findings 2025)

• MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer Decoding (Zuhri et al., Findings 2025)

ACL

• Zayd Muhammad Kawakibi Zuhri, Muhammad Farid Adilazuarda, Ayu Purwarianti, and Alham Fikri Aji. 2025. MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer Decoding. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 5516–5525, Albuquerque, New Mexico. Association for Computational Linguistics.