<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="berkane-majumder-2026-epicurvebench">
    <titleInfo>
        <title>EpiCurveBench: Evaluating VLMs on Epidemic Curve Digitization</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Thomas</namePart>
        <namePart type="family">Berkane</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Maimuna</namePart>
        <namePart type="given">S</namePart>
        <namePart type="family">Majumder</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>Chart-to-data extraction with vision-language models (VLMs) is increasingly evaluated on benchmarks that show diminishing headroom (frontier VLMs exceed 89% on ChartQA) and with metrics that treat extracted points as unordered key-value pairs, ignoring the temporal structure of time series and penalizing small alignment shifts as catastrophic failures. We address both gaps with EpiCurveBench, a benchmark of 1,000 epidemic curve images curated from real-world public-health sources, and EpiCurveSimilarity (ECS), an evaluation metric that aligns predicted and ground-truth series via dynamic programming, tolerating local temporal shifts and gaps while penalizing them proportionally. Evaluating nine methods—three frontier closed VLMs, four open VLM, and two specialized chart-extraction systems—we find the strongest model reaches only 52.3% ECS, and that ECS spreads the four general-purpose VLMs over a 25-point range where key-value metrics (RMS, SCRM) compress them into a 5-point band. We further validate ECS against four downstream epidemiological summary statistics, finding that higher ECS predicts smaller errors in total counts, peak timing, and peak magnitude, and higher growth-rate fidelity; across all four, ECS correlates 1.5–3.6\times more strongly than Dynamic Time Warping, which lacks a gap penalty and therefore cannot distinguish a truncated prediction from a temporally faithful one. EpiCurveBench targets a high-impact public-health application—unlocking decades of outbreak data trapped in published figures—but the benchmark and metric apply directly to any structured time-series chart-extraction setting.</abstract>
    <identifier type="citekey">berkane-majumder-2026-epicurvebench</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1409/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>32188</start>
            <end>32201</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T EpiCurveBench: Evaluating VLMs on Epidemic Curve Digitization
%A Berkane, Thomas
%A Majumder, Maimuna S.
%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 berkane-majumder-2026-epicurvebench
%X Chart-to-data extraction with vision-language models (VLMs) is increasingly evaluated on benchmarks that show diminishing headroom (frontier VLMs exceed 89% on ChartQA) and with metrics that treat extracted points as unordered key-value pairs, ignoring the temporal structure of time series and penalizing small alignment shifts as catastrophic failures. We address both gaps with EpiCurveBench, a benchmark of 1,000 epidemic curve images curated from real-world public-health sources, and EpiCurveSimilarity (ECS), an evaluation metric that aligns predicted and ground-truth series via dynamic programming, tolerating local temporal shifts and gaps while penalizing them proportionally. Evaluating nine methods—three frontier closed VLMs, four open VLM, and two specialized chart-extraction systems—we find the strongest model reaches only 52.3% ECS, and that ECS spreads the four general-purpose VLMs over a 25-point range where key-value metrics (RMS, SCRM) compress them into a 5-point band. We further validate ECS against four downstream epidemiological summary statistics, finding that higher ECS predicts smaller errors in total counts, peak timing, and peak magnitude, and higher growth-rate fidelity; across all four, ECS correlates 1.5–3.6\times more strongly than Dynamic Time Warping, which lacks a gap penalty and therefore cannot distinguish a truncated prediction from a temporally faithful one. EpiCurveBench targets a high-impact public-health application—unlocking decades of outbreak data trapped in published figures—but the benchmark and metric apply directly to any structured time-series chart-extraction setting.
%U https://aclanthology.org/2026.emnlp-main.1409/
%P 32188-32201

Markdown (Informal)

[EpiCurveBench: Evaluating VLMs on Epidemic Curve Digitization](https://aclanthology.org/2026.emnlp-main.1409/) (Berkane & Majumder, EMNLP 2026)

• EpiCurveBench: Evaluating VLMs on Epidemic Curve Digitization (Berkane & Majumder, EMNLP 2026)

ACL

• Thomas Berkane and Maimuna S. Majumder. 2026. EpiCurveBench: Evaluating VLMs on Epidemic Curve Digitization. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 32188–32201, Budapest, Hungary. Association for Computational Linguistics.