@inproceedings{kimhi-etal-2026-cares,
title = "{CARES}: Context-Aware Resolution Selector for {VLM}s",
author = "Kimhi, Moshe and
Shabtay, Nimrod and
Giryes, Raja and
Baskin, Chaim and
Schwartz, Eli",
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.102/",
doi = "10.18653/v1/2026.acl-long.102",
pages = "2243--2256",
ISBN = "979-8-89176-390-6",
abstract = "Large vision{--}language models (VLMs) commonly process images at native or high resolution to remain effective across tasks. This inflates visual tokens to 97-99{\%} of total tokens, resulting in high compute and latency, even when low-resolution images would suffice. We introduce \textit{CARES}{---}a \textbf{C}ontext-\textbf{A}ware \textbf{R}esolution \textbf{S}elector, a lightweight preprocessing module that, given an image{--}query pair, predicts the \textit{minimal} sufficient input resolution. CARES uses a compact VLM (350M) to extract features and predict when a target pretrained VLM{'}s response converges to its peak ability to answer correctly. Though trained as a discrete classifier over a set of optional resolutions, CARES interpolates continuous resolutions at inference for fine-grained control. Across five multimodal benchmarks spanning documents and natural images, as well as diverse target VLMs, CARES preserves task performance while reducing compute by up to 80{\%}."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kimhi-etal-2026-cares">
<titleInfo>
<title>CARES: Context-Aware Resolution Selector for VLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Moshe</namePart>
<namePart type="family">Kimhi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nimrod</namePart>
<namePart type="family">Shabtay</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raja</namePart>
<namePart type="family">Giryes</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chaim</namePart>
<namePart type="family">Baskin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eli</namePart>
<namePart type="family">Schwartz</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 vision–language models (VLMs) commonly process images at native or high resolution to remain effective across tasks. This inflates visual tokens to 97-99% of total tokens, resulting in high compute and latency, even when low-resolution images would suffice. We introduce CARES—a Context-Aware Resolution Selector, a lightweight preprocessing module that, given an image–query pair, predicts the minimal sufficient input resolution. CARES uses a compact VLM (350M) to extract features and predict when a target pretrained VLM’s response converges to its peak ability to answer correctly. Though trained as a discrete classifier over a set of optional resolutions, CARES interpolates continuous resolutions at inference for fine-grained control. Across five multimodal benchmarks spanning documents and natural images, as well as diverse target VLMs, CARES preserves task performance while reducing compute by up to 80%.</abstract>
<identifier type="citekey">kimhi-etal-2026-cares</identifier>
<identifier type="doi">10.18653/v1/2026.acl-long.102</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.102/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>2243</start>
<end>2256</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CARES: Context-Aware Resolution Selector for VLMs
%A Kimhi, Moshe
%A Shabtay, Nimrod
%A Giryes, Raja
%A Baskin, Chaim
%A Schwartz, Eli
%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 kimhi-etal-2026-cares
%X Large vision–language models (VLMs) commonly process images at native or high resolution to remain effective across tasks. This inflates visual tokens to 97-99% of total tokens, resulting in high compute and latency, even when low-resolution images would suffice. We introduce CARES—a Context-Aware Resolution Selector, a lightweight preprocessing module that, given an image–query pair, predicts the minimal sufficient input resolution. CARES uses a compact VLM (350M) to extract features and predict when a target pretrained VLM’s response converges to its peak ability to answer correctly. Though trained as a discrete classifier over a set of optional resolutions, CARES interpolates continuous resolutions at inference for fine-grained control. Across five multimodal benchmarks spanning documents and natural images, as well as diverse target VLMs, CARES preserves task performance while reducing compute by up to 80%.
%R 10.18653/v1/2026.acl-long.102
%U https://aclanthology.org/2026.acl-long.102/
%U https://doi.org/10.18653/v1/2026.acl-long.102
%P 2243-2256
Markdown (Informal)
[CARES: Context-Aware Resolution Selector for VLMs](https://aclanthology.org/2026.acl-long.102/) (Kimhi et al., ACL 2026)
ACL
- Moshe Kimhi, Nimrod Shabtay, Raja Giryes, Chaim Baskin, and Eli Schwartz. 2026. CARES: Context-Aware Resolution Selector for VLMs. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2243–2256, San Diego, California, United States. Association for Computational Linguistics.