@inproceedings{corallo-etal-2026-cachenotes,
title = "{C}ache{N}otes: Task-Aware Key-Value Cache Compression for Reasoning-Intensive Knowledge Tasks",
author = "Corallo, Giulio and
Weller, Orion and
Petroni, Fabio and
Papotti, Paolo",
editor = "Demberg, Vera and
Inui, Kentaro and
Marquez, Llu{\'i}s",
booktitle = "Proceedings of the 19th Conference of the {E}uropean Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.eacl-long.309/",
pages = "6571--6590",
ISBN = "979-8-89176-380-7",
abstract = "Integrating external knowledge into Large Language Models (LLMs) iscrucial for many real-world applications, yet current methods like Retrieval-Augmented Generation (RAG) face limitations with broad, multi-source queries, while long-context models are computationally prohibitive.We introduce CacheNotes: Task-Aware Key-Value Cache Compression. Given a task description and a corpus, CacheNotes first generates a sequence of Compression-Planning-Tokens (CPTs), an offline task-focused distillation pass that identifies and organizes key information from the corpus. These CPTs are then used to guide a one-time compression of the corpus into a compact, reusable KV cache, which is then used alone at inference time to efficiently answer diverse, reasoning-intensive queries, eliminating repeated retrieval or context expansion.Experiments on LongBench show that, on Question-Answering tasks at a $20\times$ compression, CacheNotes outperforms RAG by over 8 F1 points and reduces latency by over $4\times$. On RULER, it surpasses previous query-agnostic compression methods by 55 points, narrowing the gap to query-aware compression approaches. Additional results on real-world enterprise and synthetic datasets demonstrate its strong performance on multi-hop and broad-coverage queries."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="corallo-etal-2026-cachenotes">
<titleInfo>
<title>CacheNotes: Task-Aware Key-Value Cache Compression for Reasoning-Intensive Knowledge Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Giulio</namePart>
<namePart type="family">Corallo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Orion</namePart>
<namePart type="family">Weller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabio</namePart>
<namePart type="family">Petroni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paolo</namePart>
<namePart type="family">Papotti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vera</namePart>
<namePart type="family">Demberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lluís</namePart>
<namePart type="family">Marquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-380-7</identifier>
</relatedItem>
<abstract>Integrating external knowledge into Large Language Models (LLMs) iscrucial for many real-world applications, yet current methods like Retrieval-Augmented Generation (RAG) face limitations with broad, multi-source queries, while long-context models are computationally prohibitive.We introduce CacheNotes: Task-Aware Key-Value Cache Compression. Given a task description and a corpus, CacheNotes first generates a sequence of Compression-Planning-Tokens (CPTs), an offline task-focused distillation pass that identifies and organizes key information from the corpus. These CPTs are then used to guide a one-time compression of the corpus into a compact, reusable KV cache, which is then used alone at inference time to efficiently answer diverse, reasoning-intensive queries, eliminating repeated retrieval or context expansion.Experiments on LongBench show that, on Question-Answering tasks at a 20\times compression, CacheNotes outperforms RAG by over 8 F1 points and reduces latency by over 4\times. On RULER, it surpasses previous query-agnostic compression methods by 55 points, narrowing the gap to query-aware compression approaches. Additional results on real-world enterprise and synthetic datasets demonstrate its strong performance on multi-hop and broad-coverage queries.</abstract>
<identifier type="citekey">corallo-etal-2026-cachenotes</identifier>
<location>
<url>https://aclanthology.org/2026.eacl-long.309/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>6571</start>
<end>6590</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CacheNotes: Task-Aware Key-Value Cache Compression for Reasoning-Intensive Knowledge Tasks
%A Corallo, Giulio
%A Weller, Orion
%A Petroni, Fabio
%A Papotti, Paolo
%Y Demberg, Vera
%Y Inui, Kentaro
%Y Marquez, Lluís
%S Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-380-7
%F corallo-etal-2026-cachenotes
%X Integrating external knowledge into Large Language Models (LLMs) iscrucial for many real-world applications, yet current methods like Retrieval-Augmented Generation (RAG) face limitations with broad, multi-source queries, while long-context models are computationally prohibitive.We introduce CacheNotes: Task-Aware Key-Value Cache Compression. Given a task description and a corpus, CacheNotes first generates a sequence of Compression-Planning-Tokens (CPTs), an offline task-focused distillation pass that identifies and organizes key information from the corpus. These CPTs are then used to guide a one-time compression of the corpus into a compact, reusable KV cache, which is then used alone at inference time to efficiently answer diverse, reasoning-intensive queries, eliminating repeated retrieval or context expansion.Experiments on LongBench show that, on Question-Answering tasks at a 20\times compression, CacheNotes outperforms RAG by over 8 F1 points and reduces latency by over 4\times. On RULER, it surpasses previous query-agnostic compression methods by 55 points, narrowing the gap to query-aware compression approaches. Additional results on real-world enterprise and synthetic datasets demonstrate its strong performance on multi-hop and broad-coverage queries.
%U https://aclanthology.org/2026.eacl-long.309/
%P 6571-6590
Markdown (Informal)
[CacheNotes: Task-Aware Key-Value Cache Compression for Reasoning-Intensive Knowledge Tasks](https://aclanthology.org/2026.eacl-long.309/) (Corallo et al., EACL 2026)
ACL