@inproceedings{xue-etal-2026-benchmarking,
title = "Benchmarking and Mitigating the Impact of Noisy User Prompts in Medical {VLM}s via Cross-Modal Reflection",
author = "Xue, Zhiyu and
Abbasi-Asl, Reza and
Pedarsani, Ramtin",
editor = {Matusevych, Yevgen and
Eryi{\u{g}}it, G{\"u}l{\c{s}}en and
Aletras, Nikolaos},
booktitle = "Proceedings of the 19th Conference of the {E}uropean Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 5: Industry Track)",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.eacl-industry.67/",
pages = "900--914",
ISBN = "979-8-89176-384-5",
abstract = "Medical vision-language models (Med-VLMs) offer a new and effective paradigm for digital health in tasks such as disease diagnosis using clinical images and text. In these tasks, an important but underexplored research question is how Med-VLMs interpret and respond to user-provided clinical information, especially when the prompts are noisy. For a systematic evaluation, we construct Med-CP, a large-scale visual question answering (VQA) benchmark designed to comprehensively evaluate the influence of clinical prompts across diverse modalities, anatomical regions, and diagnostic tasks. Our experiments reveal that existing Med-VLMs tend to follow user-provided prompts blindly, regardless of whether they are accurate or not, raising concerns about their reliability in real-world interactions. To address this problem, we introduce a novel supervised fine-tuning (SFT) approach for Med-VLMs based on cross-modal reflection chain-of-thought (CoT) across medical images and text. In our SFT method, the Med-VLM is trained to produce reasoning paths for the analysis of the medical image and the user-provided prompt. Then, the final answer is determined by conducting a reflection on the visual and textual information. Experimental results demonstrate that our method considerably enhances the robustness against noisy user-provided prompts for both in-domain and out-of-domain evaluation scenarios."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xue-etal-2026-benchmarking">
<titleInfo>
<title>Benchmarking and Mitigating the Impact of Noisy User Prompts in Medical VLMs via Cross-Modal Reflection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhiyu</namePart>
<namePart type="family">Xue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Reza</namePart>
<namePart type="family">Abbasi-Asl</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ramtin</namePart>
<namePart type="family">Pedarsani</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 5: Industry Track)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yevgen</namePart>
<namePart type="family">Matusevych</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gülşen</namePart>
<namePart type="family">Eryiğit</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolaos</namePart>
<namePart type="family">Aletras</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">979-8-89176-384-5</identifier>
</relatedItem>
<abstract>Medical vision-language models (Med-VLMs) offer a new and effective paradigm for digital health in tasks such as disease diagnosis using clinical images and text. In these tasks, an important but underexplored research question is how Med-VLMs interpret and respond to user-provided clinical information, especially when the prompts are noisy. For a systematic evaluation, we construct Med-CP, a large-scale visual question answering (VQA) benchmark designed to comprehensively evaluate the influence of clinical prompts across diverse modalities, anatomical regions, and diagnostic tasks. Our experiments reveal that existing Med-VLMs tend to follow user-provided prompts blindly, regardless of whether they are accurate or not, raising concerns about their reliability in real-world interactions. To address this problem, we introduce a novel supervised fine-tuning (SFT) approach for Med-VLMs based on cross-modal reflection chain-of-thought (CoT) across medical images and text. In our SFT method, the Med-VLM is trained to produce reasoning paths for the analysis of the medical image and the user-provided prompt. Then, the final answer is determined by conducting a reflection on the visual and textual information. Experimental results demonstrate that our method considerably enhances the robustness against noisy user-provided prompts for both in-domain and out-of-domain evaluation scenarios.</abstract>
<identifier type="citekey">xue-etal-2026-benchmarking</identifier>
<location>
<url>https://aclanthology.org/2026.eacl-industry.67/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>900</start>
<end>914</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Benchmarking and Mitigating the Impact of Noisy User Prompts in Medical VLMs via Cross-Modal Reflection
%A Xue, Zhiyu
%A Abbasi-Asl, Reza
%A Pedarsani, Ramtin
%Y Matusevych, Yevgen
%Y Eryiğit, Gülşen
%Y Aletras, Nikolaos
%S Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 5: Industry Track)
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-384-5
%F xue-etal-2026-benchmarking
%X Medical vision-language models (Med-VLMs) offer a new and effective paradigm for digital health in tasks such as disease diagnosis using clinical images and text. In these tasks, an important but underexplored research question is how Med-VLMs interpret and respond to user-provided clinical information, especially when the prompts are noisy. For a systematic evaluation, we construct Med-CP, a large-scale visual question answering (VQA) benchmark designed to comprehensively evaluate the influence of clinical prompts across diverse modalities, anatomical regions, and diagnostic tasks. Our experiments reveal that existing Med-VLMs tend to follow user-provided prompts blindly, regardless of whether they are accurate or not, raising concerns about their reliability in real-world interactions. To address this problem, we introduce a novel supervised fine-tuning (SFT) approach for Med-VLMs based on cross-modal reflection chain-of-thought (CoT) across medical images and text. In our SFT method, the Med-VLM is trained to produce reasoning paths for the analysis of the medical image and the user-provided prompt. Then, the final answer is determined by conducting a reflection on the visual and textual information. Experimental results demonstrate that our method considerably enhances the robustness against noisy user-provided prompts for both in-domain and out-of-domain evaluation scenarios.
%U https://aclanthology.org/2026.eacl-industry.67/
%P 900-914
Markdown (Informal)
[Benchmarking and Mitigating the Impact of Noisy User Prompts in Medical VLMs via Cross-Modal Reflection](https://aclanthology.org/2026.eacl-industry.67/) (Xue et al., EACL 2026)
ACL