@inproceedings{zhou-etal-2026-scvq,
title = "{SCVQ}: Sparse-Compensated Vector Quantization for Large Language Models",
author = "Zhou, Zixuan and
Diao, Yujun and
Kong, Zicheng and
Ma, Dehua and
Xu, Zhenbo and
Li, Pei Pei and
He, Zhaofeng",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.403/",
pages = "8934--8950",
ISBN = "979-8-89176-390-6",
abstract = "Large Language Models (LLMs) are primarily constrained by memory and bandwidth bottlenecks during deployment. Although Vector Quantization (VQ) has emerged as a promising solution, existing methods incur inference overhead due to massive codebook storage and intensive index lookups. Moreover, these methods typically suffer from non-negligible performance degradation under ultra-low bitwidth regimes. To bridge this gap, we propose Sparse-Compensated Vector Quantization (SCVQ), a novel framework designed for high-efficiency LLM vector quantization. SCVQ introduces a salience-aware weighted K-means clustering scheme with symmetry constraints to reduces codebook size and indexing costs. Central to our approach is a unified structured representation that consolidates outliers, salient weights, and quantization residuals into a single sparse compensation matrix. This design effectively preserves critical model information while leveraging VQ-specific properties to enable efficient custom kernels. Extensive experiments across multiple benchmarks demonstrate SCVQ{'}s superior performance. Specifically, SCVQ achieves a perplexity of 5.78 on WikiText-2 for LLaMA-2-7B at 2-bit quantization, while delivering a $1.4\times$ end-to-end inference speedup over existing baselines."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhou-etal-2026-scvq">
<titleInfo>
<title>SCVQ: Sparse-Compensated Vector Quantization for Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zixuan</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yujun</namePart>
<namePart type="family">Diao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zicheng</namePart>
<namePart type="family">Kong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dehua</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhenbo</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pei</namePart>
<namePart type="given">Pei</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhaofeng</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) are primarily constrained by memory and bandwidth bottlenecks during deployment. Although Vector Quantization (VQ) has emerged as a promising solution, existing methods incur inference overhead due to massive codebook storage and intensive index lookups. Moreover, these methods typically suffer from non-negligible performance degradation under ultra-low bitwidth regimes. To bridge this gap, we propose Sparse-Compensated Vector Quantization (SCVQ), a novel framework designed for high-efficiency LLM vector quantization. SCVQ introduces a salience-aware weighted K-means clustering scheme with symmetry constraints to reduces codebook size and indexing costs. Central to our approach is a unified structured representation that consolidates outliers, salient weights, and quantization residuals into a single sparse compensation matrix. This design effectively preserves critical model information while leveraging VQ-specific properties to enable efficient custom kernels. Extensive experiments across multiple benchmarks demonstrate SCVQ’s superior performance. Specifically, SCVQ achieves a perplexity of 5.78 on WikiText-2 for LLaMA-2-7B at 2-bit quantization, while delivering a 1.4\times end-to-end inference speedup over existing baselines.</abstract>
<identifier type="citekey">zhou-etal-2026-scvq</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.403/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>8934</start>
<end>8950</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SCVQ: Sparse-Compensated Vector Quantization for Large Language Models
%A Zhou, Zixuan
%A Diao, Yujun
%A Kong, Zicheng
%A Ma, Dehua
%A Xu, Zhenbo
%A Li, Pei Pei
%A He, Zhaofeng
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F zhou-etal-2026-scvq
%X Large Language Models (LLMs) are primarily constrained by memory and bandwidth bottlenecks during deployment. Although Vector Quantization (VQ) has emerged as a promising solution, existing methods incur inference overhead due to massive codebook storage and intensive index lookups. Moreover, these methods typically suffer from non-negligible performance degradation under ultra-low bitwidth regimes. To bridge this gap, we propose Sparse-Compensated Vector Quantization (SCVQ), a novel framework designed for high-efficiency LLM vector quantization. SCVQ introduces a salience-aware weighted K-means clustering scheme with symmetry constraints to reduces codebook size and indexing costs. Central to our approach is a unified structured representation that consolidates outliers, salient weights, and quantization residuals into a single sparse compensation matrix. This design effectively preserves critical model information while leveraging VQ-specific properties to enable efficient custom kernels. Extensive experiments across multiple benchmarks demonstrate SCVQ’s superior performance. Specifically, SCVQ achieves a perplexity of 5.78 on WikiText-2 for LLaMA-2-7B at 2-bit quantization, while delivering a 1.4\times end-to-end inference speedup over existing baselines.
%U https://aclanthology.org/2026.acl-long.403/
%P 8934-8950
Markdown (Informal)
[SCVQ: Sparse-Compensated Vector Quantization for Large Language Models](https://aclanthology.org/2026.acl-long.403/) (Zhou et al., ACL 2026)
ACL
- Zixuan Zhou, Yujun Diao, Zicheng Kong, Dehua Ma, Zhenbo Xu, Pei Pei Li, and Zhaofeng He. 2026. SCVQ: Sparse-Compensated Vector Quantization for Large Language Models. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8934–8950, San Diego, California, United States. Association for Computational Linguistics.