<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xu-etal-2026-blueprintagent">
    <titleInfo>
        <title>BlueprintAgent: Constraint-Triggered Targeted Revisits for Simulation-Ready Generation from Scanned Structural Blueprints</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Zhouyuan</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Chen</namePart>
        <namePart type="family">Yang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Linhao</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiansheng</namePart>
        <namePart type="family">Fan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Chen</namePart>
        <namePart type="family">Wang</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>Converting in-service reinforced-concrete (RC) building blueprints into simulation-ready models underpins safety assessment and seismic retrofit, but the process remains largely manual. Direct prompting of a multimodal large language model (MLLM) over a full scanned sheet is unreliable: plausible-looking outputs often violate engineering constraints on beam–column support, span count, or 3D continuity.We present BlueprintAgent (BPA), a constraint-triggered multimodal agent for simulation-ready frame extraction from scanned blueprints. BPA treats the MLLM as the primary reader and decision maker, with OCR and computer vision supplying localized evidence. Its novelty is not the tool stack but a control mechanism: engineering constraints are realized as callable validators whose entity-level conflict reports trigger targeted MLLM revisits over the relevant local region—an inference-time control distinct from fixed pipelines and free-form self-reflection.We evaluate BPA on 300 real scanned blueprint sheets from 20 anonymized RC frame projects, against five baselines and six ablations. BPA reaches a macro-averaged Beam F1 of 0.994, against 0.301 for single-MLLM zero-shot and 0.820 for a fixed pipeline; removing MLLM-led axis adjudication collapses Beam and Column F1 on complex multi-sheet projects. For dense technical drawings, engineering constraints are best deployed as triggers for entity-level targeted revisits rather than as post-hoc output filters.</abstract>
    <identifier type="citekey">xu-etal-2026-blueprintagent</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.634/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>13304</start>
            <end>13317</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T BlueprintAgent: Constraint-Triggered Targeted Revisits for Simulation-Ready Generation from Scanned Structural Blueprints
%A Xu, Zhouyuan
%A Yang, Chen
%A Wang, Linhao
%A Fan, Jiansheng
%A Wang, Chen
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F xu-etal-2026-blueprintagent
%X Converting in-service reinforced-concrete (RC) building blueprints into simulation-ready models underpins safety assessment and seismic retrofit, but the process remains largely manual. Direct prompting of a multimodal large language model (MLLM) over a full scanned sheet is unreliable: plausible-looking outputs often violate engineering constraints on beam–column support, span count, or 3D continuity.We present BlueprintAgent (BPA), a constraint-triggered multimodal agent for simulation-ready frame extraction from scanned blueprints. BPA treats the MLLM as the primary reader and decision maker, with OCR and computer vision supplying localized evidence. Its novelty is not the tool stack but a control mechanism: engineering constraints are realized as callable validators whose entity-level conflict reports trigger targeted MLLM revisits over the relevant local region—an inference-time control distinct from fixed pipelines and free-form self-reflection.We evaluate BPA on 300 real scanned blueprint sheets from 20 anonymized RC frame projects, against five baselines and six ablations. BPA reaches a macro-averaged Beam F1 of 0.994, against 0.301 for single-MLLM zero-shot and 0.820 for a fixed pipeline; removing MLLM-led axis adjudication collapses Beam and Column F1 on complex multi-sheet projects. For dense technical drawings, engineering constraints are best deployed as triggers for entity-level targeted revisits rather than as post-hoc output filters.
%U https://aclanthology.org/2026.findings-emnlp.634/
%P 13304-13317

Markdown (Informal)

[BlueprintAgent: Constraint-Triggered Targeted Revisits for Simulation-Ready Generation from Scanned Structural Blueprints](https://aclanthology.org/2026.findings-emnlp.634/) (Xu et al., Findings 2026)

• BlueprintAgent: Constraint-Triggered Targeted Revisits for Simulation-Ready Generation from Scanned Structural Blueprints (Xu et al., Findings 2026)

ACL

• Zhouyuan Xu, Chen Yang, Linhao Wang, Jiansheng Fan, and Chen Wang. 2026. BlueprintAgent: Constraint-Triggered Targeted Revisits for Simulation-Ready Generation from Scanned Structural Blueprints. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 13304–13317, Budapest, Hungary. Association for Computational Linguistics.