<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ma-etal-2026-fluid">
    <titleInfo>
        <title>FLUID-V: Elastic-Horizon Diffusion Decoding for Vision-Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Xiangyu</namePart>
        <namePart type="family">Ma</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Teng</namePart>
        <namePart type="family">Xiao</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>
    <name type="personal">
        <namePart type="given">Ping</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hai</namePart>
        <namePart type="family">Zhao</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>Autoregressive vision-language models achieve strong multimodal understanding, but their sequential decoding becomes increasingly inefficient as multimodal responses grow longer. Adapting AR-VLMs to diffusion-style block decoding is promising, yet fixed refinement blocks struggle with sharply varying multimodal uncertainty in OCR, chart, document, and reasoning-heavy outputs. Binary mask-or-token states further limit local refinement, as masking discards partial evidence while premature commitment can corrupt the causal prefix. We present FLUID-V, a lightweight AR-to-diffusion adaptation framework for reliable local multimodal refinement. FLUID-V selects refinement spans with elastic horizons and represents unresolved positions as evolving soft semantic mixtures in embedding space. Only confident predictions are committed to the persistent KV cache, preserving causal multimodal conditioning and autoregressive compatibility. Experiments with OpenPangu-VL-7B and 100K self-distilled samples show competitive performance across diverse multimodal benchmarks, with ablations confirming the importance of adaptive horizons and stateful intermediate representations. Our code is available at https://github.com/Oli-lab-nun/FLUID-V.</abstract>
    <identifier type="citekey">ma-etal-2026-fluid</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.347/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>7990</start>
            <end>8003</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T FLUID-V: Elastic-Horizon Diffusion Decoding for Vision-Language Models
%A Ma, Xiangyu
%A Xiao, Teng
%A Li, Zuchao
%A Wang, Ping
%A Zhao, Hai
%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 ma-etal-2026-fluid
%X Autoregressive vision-language models achieve strong multimodal understanding, but their sequential decoding becomes increasingly inefficient as multimodal responses grow longer. Adapting AR-VLMs to diffusion-style block decoding is promising, yet fixed refinement blocks struggle with sharply varying multimodal uncertainty in OCR, chart, document, and reasoning-heavy outputs. Binary mask-or-token states further limit local refinement, as masking discards partial evidence while premature commitment can corrupt the causal prefix. We present FLUID-V, a lightweight AR-to-diffusion adaptation framework for reliable local multimodal refinement. FLUID-V selects refinement spans with elastic horizons and represents unresolved positions as evolving soft semantic mixtures in embedding space. Only confident predictions are committed to the persistent KV cache, preserving causal multimodal conditioning and autoregressive compatibility. Experiments with OpenPangu-VL-7B and 100K self-distilled samples show competitive performance across diverse multimodal benchmarks, with ablations confirming the importance of adaptive horizons and stateful intermediate representations. Our code is available at https://github.com/Oli-lab-nun/FLUID-V.
%U https://aclanthology.org/2026.emnlp-main.347/
%P 7990-8003

Markdown (Informal)

[FLUID-V: Elastic-Horizon Diffusion Decoding for Vision-Language Models](https://aclanthology.org/2026.emnlp-main.347/) (Ma et al., EMNLP 2026)

• FLUID-V: Elastic-Horizon Diffusion Decoding for Vision-Language Models (Ma et al., EMNLP 2026)

ACL

• Xiangyu Ma, Teng Xiao, Zuchao Li, Ping Wang, and Hai Zhao. 2026. FLUID-V: Elastic-Horizon Diffusion Decoding for Vision-Language Models. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 7990–8003, Budapest, Hungary. Association for Computational Linguistics.