@inproceedings{lee-etal-2025-amq,
title = "{AMQ}: Enabling {A}uto{ML} for Mixed-precision Weight-Only Quantization of Large Language Models",
author = "Lee, Sangjun and
Woo, Seung-taek and
Jin, Jun-gyu and
Lee, Changhun and
Park, Eunhyeok",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-main.1799/",
doi = "10.18653/v1/2025.emnlp-main.1799",
pages = "35532--35550",
ISBN = "979-8-89176-332-6",
abstract = "To enable broader deployment of Large Language Models (LLMs), it is essential to identify the best-performing model under strict memory constraints. We present AMQ, Automated Mixed-Precision Weight-Only Quantization, a framework that assigns layer-wise quantization bit-widths to optimally balance model quality and memory usage. However, the combinatorial search space, with over $10^{100}$ possible configurations, makes conventional black-box optimization infeasible. AMQ overcomes this challenge through four key innovations: (1) \textbf{search space pruning} using prior knowledge to exclude unpromising configurations, (2) \textbf{quantization proxy} to bypass costly format conversions during search, (3) \textbf{quality predictor} to minimize evaluation overhead, and (4) \textbf{iterative search-and-update} strategy for fast and stable convergence. By integrating these components, AMQ efficiently explores the quality{--}efficiency landscape, reaching the Pareto frontier and yielding LLMs that are both compact and high-performing."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lee-etal-2025-amq">
<titleInfo>
<title>AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sangjun</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seung-taek</namePart>
<namePart type="family">Woo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun-gyu</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Changhun</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eunhyeok</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-332-6</identifier>
</relatedItem>
<abstract>To enable broader deployment of Large Language Models (LLMs), it is essential to identify the best-performing model under strict memory constraints. We present AMQ, Automated Mixed-Precision Weight-Only Quantization, a framework that assigns layer-wise quantization bit-widths to optimally balance model quality and memory usage. However, the combinatorial search space, with over 10¹00 possible configurations, makes conventional black-box optimization infeasible. AMQ overcomes this challenge through four key innovations: (1) search space pruning using prior knowledge to exclude unpromising configurations, (2) quantization proxy to bypass costly format conversions during search, (3) quality predictor to minimize evaluation overhead, and (4) iterative search-and-update strategy for fast and stable convergence. By integrating these components, AMQ efficiently explores the quality–efficiency landscape, reaching the Pareto frontier and yielding LLMs that are both compact and high-performing.</abstract>
<identifier type="citekey">lee-etal-2025-amq</identifier>
<identifier type="doi">10.18653/v1/2025.emnlp-main.1799</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-main.1799/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>35532</start>
<end>35550</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models
%A Lee, Sangjun
%A Woo, Seung-taek
%A Jin, Jun-gyu
%A Lee, Changhun
%A Park, Eunhyeok
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F lee-etal-2025-amq
%X To enable broader deployment of Large Language Models (LLMs), it is essential to identify the best-performing model under strict memory constraints. We present AMQ, Automated Mixed-Precision Weight-Only Quantization, a framework that assigns layer-wise quantization bit-widths to optimally balance model quality and memory usage. However, the combinatorial search space, with over 10¹00 possible configurations, makes conventional black-box optimization infeasible. AMQ overcomes this challenge through four key innovations: (1) search space pruning using prior knowledge to exclude unpromising configurations, (2) quantization proxy to bypass costly format conversions during search, (3) quality predictor to minimize evaluation overhead, and (4) iterative search-and-update strategy for fast and stable convergence. By integrating these components, AMQ efficiently explores the quality–efficiency landscape, reaching the Pareto frontier and yielding LLMs that are both compact and high-performing.
%R 10.18653/v1/2025.emnlp-main.1799
%U https://aclanthology.org/2025.emnlp-main.1799/
%U https://doi.org/10.18653/v1/2025.emnlp-main.1799
%P 35532-35550
Markdown (Informal)
[AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models](https://aclanthology.org/2025.emnlp-main.1799/) (Lee et al., EMNLP 2025)
ACL