<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="qu-etal-2026-dlsa">
    <titleInfo>
        <title>DLSA: Dual-Level Semantic Alignment with Adaptive Multi-Scale Modeling for Referring Remote Sensing Image Segmentation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Wei</namePart>
        <namePart type="family">Qu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Wei</namePart>
        <namePart type="family">Wei</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Shihao</namePart>
        <namePart type="family">Zou</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zian</namePart>
        <namePart type="family">Cao</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>Referring Remote Sensing Image Segmentation (RRSIS) aims to segment the corresponding target region in a remote sensing image according to a natural-language expression. Compared with conventional referring image segmentation, RRSIS faces more significant scale variations, complex spatial layouts, and visual distractions, placing higher demands on vision–language alignment and multi-scale modeling. To address these challenges, we propose DLSA, a dual-level semantic alignment framework for RRSIS. DLSA introduces semantic alignment at two complementary levels, namely encoding and decoding. At the encoder level, we design Semantic-guided Adaptive Multi-scale Convolution (SgAMC), which explicitly introduces linguistic semantics into receptive-field selection, enabling language-guided adaptive scale modeling. At the decoder level, we introduce Adaptive Late-stage Text Injection (ALTI), which adaptively injects textual semantics into late decoding stages to refine mask reconstruction while preserving the stability of the decoding pathway. Experiments on RRSIS-D and RefSegRS demonstrate that DLSA achieves strong performance, validating the effectiveness of dual-level semantic alignment in remote sensing image scenarios. The source code is publicly available at https://github.com/young-qkl/DLSA.</abstract>
    <identifier type="citekey">qu-etal-2026-dlsa</identifier>
    <identifier type="doi">10.18653/2026.findings-emnlp.1026</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.1026/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>21687</start>
            <end>21704</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T DLSA: Dual-Level Semantic Alignment with Adaptive Multi-Scale Modeling for Referring Remote Sensing Image Segmentation
%A Qu, Wei
%A Wei, Wei
%A Zou, Shihao
%A Cao, Zian
%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 qu-etal-2026-dlsa
%X Referring Remote Sensing Image Segmentation (RRSIS) aims to segment the corresponding target region in a remote sensing image according to a natural-language expression. Compared with conventional referring image segmentation, RRSIS faces more significant scale variations, complex spatial layouts, and visual distractions, placing higher demands on vision–language alignment and multi-scale modeling. To address these challenges, we propose DLSA, a dual-level semantic alignment framework for RRSIS. DLSA introduces semantic alignment at two complementary levels, namely encoding and decoding. At the encoder level, we design Semantic-guided Adaptive Multi-scale Convolution (SgAMC), which explicitly introduces linguistic semantics into receptive-field selection, enabling language-guided adaptive scale modeling. At the decoder level, we introduce Adaptive Late-stage Text Injection (ALTI), which adaptively injects textual semantics into late decoding stages to refine mask reconstruction while preserving the stability of the decoding pathway. Experiments on RRSIS-D and RefSegRS demonstrate that DLSA achieves strong performance, validating the effectiveness of dual-level semantic alignment in remote sensing image scenarios. The source code is publicly available at https://github.com/young-qkl/DLSA.
%R 10.18653/2026.findings-emnlp.1026
%U https://aclanthology.org/2026.findings-emnlp.1026/
%U https://doi.org/10.18653/2026.findings-emnlp.1026
%P 21687-21704

Markdown (Informal)

[DLSA: Dual-Level Semantic Alignment with Adaptive Multi-Scale Modeling for Referring Remote Sensing Image Segmentation](https://aclanthology.org/2026.findings-emnlp.1026/) (Qu et al., Findings 2026)

• DLSA: Dual-Level Semantic Alignment with Adaptive Multi-Scale Modeling for Referring Remote Sensing Image Segmentation (Qu et al., Findings 2026)

ACL

• Wei Qu, Wei Wei, Shihao Zou, and Zian Cao. 2026. DLSA: Dual-Level Semantic Alignment with Adaptive Multi-Scale Modeling for Referring Remote Sensing Image Segmentation. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 21687–21704, Budapest, Hungary. Association for Computational Linguistics.