@inproceedings{chen-etal-2026-contrastkv,
title = "{C}ontrast{KV}: Robust {KV} Cache Eviction via Contrastive Signal Fusion for Multi-Query Generalization",
author = "Chen, Xingchi and
Zong, Peiyuan and
Gao, Ziqiang and
Li, Qing and
Jiang, Yong and
Zhu, Fa and
Li, Hui",
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.417/",
pages = "9216--9229",
ISBN = "979-8-89176-390-6",
abstract = "Large Language Models (LLMs) face significant memory and latency overheads during long-context inference due to the growing KV cache, especially in Knowledge Base Question Answering (KBQA) settings that require support for multiple downstream queries. Query-aware eviction methods do not generalize across queries, while existing query-agnostic approaches rely on a single proxy query, leading to fragile eviction decisions under high eviction ratios. We propose ContrastKV, a robust query-agnostic KV cache eviction algorithm for multi-query generalization. ContrastKV introduces a contrastive signal fusion mechanism that jointly exploits complementary semantic and non-semantic signals. By contrasting semantic consistency with structural robustness, the method constructs a more reliable eviction criterion that alleviates the blind spots of single-query proxies. The framework integrates efficient signal generation, parallel importance scoring, and multi-level fusion across heads and layers. Experiments show that ContrastKV outperforms state-of-the-art methods, retaining up to 92{\%} accuracy with only 20{\%} of the KV cache budget, while reducing decoding latency by approximately 50{\%} and significantly lowering GPU memory usage."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chen-etal-2026-contrastkv">
<titleInfo>
<title>ContrastKV: Robust KV Cache Eviction via Contrastive Signal Fusion for Multi-Query Generalization</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xingchi</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peiyuan</namePart>
<namePart type="family">Zong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ziqiang</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qing</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yong</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fa</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hui</namePart>
<namePart type="family">Li</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) face significant memory and latency overheads during long-context inference due to the growing KV cache, especially in Knowledge Base Question Answering (KBQA) settings that require support for multiple downstream queries. Query-aware eviction methods do not generalize across queries, while existing query-agnostic approaches rely on a single proxy query, leading to fragile eviction decisions under high eviction ratios. We propose ContrastKV, a robust query-agnostic KV cache eviction algorithm for multi-query generalization. ContrastKV introduces a contrastive signal fusion mechanism that jointly exploits complementary semantic and non-semantic signals. By contrasting semantic consistency with structural robustness, the method constructs a more reliable eviction criterion that alleviates the blind spots of single-query proxies. The framework integrates efficient signal generation, parallel importance scoring, and multi-level fusion across heads and layers. Experiments show that ContrastKV outperforms state-of-the-art methods, retaining up to 92% accuracy with only 20% of the KV cache budget, while reducing decoding latency by approximately 50% and significantly lowering GPU memory usage.</abstract>
<identifier type="citekey">chen-etal-2026-contrastkv</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.417/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>9216</start>
<end>9229</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ContrastKV: Robust KV Cache Eviction via Contrastive Signal Fusion for Multi-Query Generalization
%A Chen, Xingchi
%A Zong, Peiyuan
%A Gao, Ziqiang
%A Li, Qing
%A Jiang, Yong
%A Zhu, Fa
%A Li, Hui
%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 chen-etal-2026-contrastkv
%X Large Language Models (LLMs) face significant memory and latency overheads during long-context inference due to the growing KV cache, especially in Knowledge Base Question Answering (KBQA) settings that require support for multiple downstream queries. Query-aware eviction methods do not generalize across queries, while existing query-agnostic approaches rely on a single proxy query, leading to fragile eviction decisions under high eviction ratios. We propose ContrastKV, a robust query-agnostic KV cache eviction algorithm for multi-query generalization. ContrastKV introduces a contrastive signal fusion mechanism that jointly exploits complementary semantic and non-semantic signals. By contrasting semantic consistency with structural robustness, the method constructs a more reliable eviction criterion that alleviates the blind spots of single-query proxies. The framework integrates efficient signal generation, parallel importance scoring, and multi-level fusion across heads and layers. Experiments show that ContrastKV outperforms state-of-the-art methods, retaining up to 92% accuracy with only 20% of the KV cache budget, while reducing decoding latency by approximately 50% and significantly lowering GPU memory usage.
%U https://aclanthology.org/2026.acl-long.417/
%P 9216-9229
Markdown (Informal)
[ContrastKV: Robust KV Cache Eviction via Contrastive Signal Fusion for Multi-Query Generalization](https://aclanthology.org/2026.acl-long.417/) (Chen et al., ACL 2026)
ACL
- Xingchi Chen, Peiyuan Zong, Ziqiang Gao, Qing Li, Yong Jiang, Fa Zhu, and Hui Li. 2026. ContrastKV: Robust KV Cache Eviction via Contrastive Signal Fusion for Multi-Query Generalization. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 9216–9229, San Diego, California, United States. Association for Computational Linguistics.