<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="he-etal-2026-multimodal">
    <titleInfo>
        <title>Multimodal Ancient Document Parsing: Technical Report for EvaHan2026 Competition</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Liqi</namePart>
        <namePart type="family">He</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Qiwei</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ziye</namePart>
        <namePart type="family">Yang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zuchao</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-05</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Fourth Workshop on Language Technologies for Historical and Ancient Languages (LT4HALA 2026) @ LREC 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Rachele</namePart>
            <namePart type="family">Sprugnoli</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Marco</namePart>
            <namePart type="family">Passarotti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>ELRA Language Resources Association (ELRA)</publisher>
            <place>
                <placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>We present the multimodal Optical Character Recognition (OCR) and layout analysis methods developed for the EvaHan 2026 competition. Our approach is built upon the Qwen2.5-VL-7B-Instruct architecture and integrates two core strategies: (1) a reinforcement learning alignment pipeline utilizing Direct Preference Optimization (DPO) and Group Relative Policy Optimization (GRPO) to explicitly mitigate hallucination and coordinate instability; and (2) a four-stage curriculum learning framework that synthesizes domain-specific historical artifacts to enhance open-modality generalization. Using this approach, we achieve competitive results, notably reaching a Character Error Rate (CER) of 0.0303 on printed texts (Task A) and 0.0552 on handwritten manuscripts (Task C), as well as an Average Intersection over Union (IoU) of 0.7638 on layout element analysis (Task B).</abstract>
    <identifier type="citekey">he-etal-2026-multimodal</identifier>
    <identifier type="doi">10.63317/2cfum2ozgjrs</identifier>
    <location>
        <url>https://aclanthology.org/2026.lt4hala-1.33/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>322</start>
            <end>329</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Multimodal Ancient Document Parsing: Technical Report for EvaHan2026 Competition
%A He, Liqi
%A Li, Qiwei
%A Yang, Ziye
%A Li, Zuchao
%Y Sprugnoli, Rachele
%Y Passarotti, Marco
%S Proceedings of the Fourth Workshop on Language Technologies for Historical and Ancient Languages (LT4HALA 2026) @ LREC 2026
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F he-etal-2026-multimodal
%X We present the multimodal Optical Character Recognition (OCR) and layout analysis methods developed for the EvaHan 2026 competition. Our approach is built upon the Qwen2.5-VL-7B-Instruct architecture and integrates two core strategies: (1) a reinforcement learning alignment pipeline utilizing Direct Preference Optimization (DPO) and Group Relative Policy Optimization (GRPO) to explicitly mitigate hallucination and coordinate instability; and (2) a four-stage curriculum learning framework that synthesizes domain-specific historical artifacts to enhance open-modality generalization. Using this approach, we achieve competitive results, notably reaching a Character Error Rate (CER) of 0.0303 on printed texts (Task A) and 0.0552 on handwritten manuscripts (Task C), as well as an Average Intersection over Union (IoU) of 0.7638 on layout element analysis (Task B).
%R 10.63317/2cfum2ozgjrs
%U https://aclanthology.org/2026.lt4hala-1.33/
%U https://doi.org/10.63317/2cfum2ozgjrs
%P 322-329

Markdown (Informal)

[Multimodal Ancient Document Parsing: Technical Report for EvaHan2026 Competition](https://aclanthology.org/2026.lt4hala-1.33/) (He et al., LT4HALA 2026)

• Multimodal Ancient Document Parsing: Technical Report for EvaHan2026 Competition (He et al., LT4HALA 2026)

ACL

• Liqi He, Qiwei Li, Ziye Yang, and Zuchao Li. 2026. Multimodal Ancient Document Parsing: Technical Report for EvaHan2026 Competition. In Proceedings of the Fourth Workshop on Language Technologies for Historical and Ancient Languages (LT4HALA 2026) @ LREC 2026, pages 322–329, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).