<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cheng-etal-2026-robust">
    <titleInfo>
        <title>How Robust is OCR-Reasoning? Evaluating OCR-Reasoning Robustness of Vision-Language Models under Visual Perturbations</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yuxing</namePart>
        <namePart type="family">Cheng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yuan</namePart>
        <namePart type="family">Wu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yi</namePart>
        <namePart type="family">Chang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Vision-language models (VLMs) achieved strong performance on OCR-based benchmarks and increasingly focused on text-rich understanding, but their robustness under controlled visual degradation remains insufficiently understood. This gap is especially critical for OCR reasoning, where visual corruption can induce OCR errors and structural distortions, thereby introducing uncertainty into reasoning task. To systematically study this problem, we introduce OCR-Robust, a benchmark designed for evaluating OCR reasoning robustness under visual perturbations. It contains 812 samples across two complementary subsets: OCR1.0, covering documents, scene text, receipts, handwriting, and mathematical content, and OCR2.0, focusing on charts, geometry diagrams, and tables. To enable efficient yet informative evaluation, we conduct a pilot study over 18 candidate perturbations and select 5 representative types at 3 severity levels each based on their impact and cross-model discriminability. We evaluate robustness using clean accuracy, Relative Corruption Retention (RCR), Worst-Case Retention (WCR), and a composite Corruption Robustness Index (CRI), and benchmark 18 models spanning proprietary systems, open-source VLMs, and OCR+LLM pipelines. Our results show that higher clean accuracy does not necessarily imply stronger robustness, and that models can suffer pronounced degradation in the worst case on OCR tasks that are sensitive to structure, and charts and tables are substantially more fragile than document-like inputs under perturbation.</abstract>
    <identifier type="citekey">cheng-etal-2026-robust</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.226/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>5220</start>
            <end>5238</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T How Robust is OCR-Reasoning? Evaluating OCR-Reasoning Robustness of Vision-Language Models under Visual Perturbations
%A Cheng, Yuxing
%A Wu, Yuan
%A Chang, Yi
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F cheng-etal-2026-robust
%X Vision-language models (VLMs) achieved strong performance on OCR-based benchmarks and increasingly focused on text-rich understanding, but their robustness under controlled visual degradation remains insufficiently understood. This gap is especially critical for OCR reasoning, where visual corruption can induce OCR errors and structural distortions, thereby introducing uncertainty into reasoning task. To systematically study this problem, we introduce OCR-Robust, a benchmark designed for evaluating OCR reasoning robustness under visual perturbations. It contains 812 samples across two complementary subsets: OCR1.0, covering documents, scene text, receipts, handwriting, and mathematical content, and OCR2.0, focusing on charts, geometry diagrams, and tables. To enable efficient yet informative evaluation, we conduct a pilot study over 18 candidate perturbations and select 5 representative types at 3 severity levels each based on their impact and cross-model discriminability. We evaluate robustness using clean accuracy, Relative Corruption Retention (RCR), Worst-Case Retention (WCR), and a composite Corruption Robustness Index (CRI), and benchmark 18 models spanning proprietary systems, open-source VLMs, and OCR+LLM pipelines. Our results show that higher clean accuracy does not necessarily imply stronger robustness, and that models can suffer pronounced degradation in the worst case on OCR tasks that are sensitive to structure, and charts and tables are substantially more fragile than document-like inputs under perturbation.
%U https://aclanthology.org/2026.emnlp-main.226/
%P 5220-5238

Markdown (Informal)

[How Robust is OCR-Reasoning? Evaluating OCR-Reasoning Robustness of Vision-Language Models under Visual Perturbations](https://aclanthology.org/2026.emnlp-main.226/) (Cheng et al., EMNLP 2026)

• How Robust is OCR-Reasoning? Evaluating OCR-Reasoning Robustness of Vision-Language Models under Visual Perturbations (Cheng et al., EMNLP 2026)

ACL

• Yuxing Cheng, Yuan Wu, and Yi Chang. 2026. How Robust is OCR-Reasoning? Evaluating OCR-Reasoning Robustness of Vision-Language Models under Visual Perturbations. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 5220–5238, Budapest, Hungary. Association for Computational Linguistics.