<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-zaiane-2026-diving">
    <titleInfo>
        <title>Diving into the Decoding Space of Non-Autoregressive Models via Lexically Constrained Search</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Chenyang</namePart>
        <namePart type="family">Huang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Osmar</namePart>
        <namePart type="family">Zaiane</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Maria</namePart>
            <namePart type="family">Liakata</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Viviane</namePart>
            <namePart type="given">P</namePart>
            <namePart type="family">Moreira</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jiajun</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">David</namePart>
            <namePart type="family">Jurgens</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">San Diego, California, United States</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-391-3</identifier>
    </relatedItem>
    <abstract>Non-autoregressive (NAR) models have been mainly developed to improve decoding efficiency. Lately, they have also shown great potential in controlled text generation tasks. In this work, we investigate the decoding space of NAR models through lexically constrained machine translation tasks, and develop a search-based decoding algorithm named LexMAP, which is comparable to the autoregressive Grid Beam Search (GBS) method. Our analysis reveals several interesting properties of NAR decoding: 1) the NAR-based method does not suffer from the MAP degradation issue as the autoregressive method does; 2) AR beam search exhibits strong positional bias, in which the candidates only diverge at the end of the sequence; 3) NAR search explores a larger portion of the probability space, suggesting that the search algorithm better exploits the model’s potential.</abstract>
    <identifier type="citekey">huang-zaiane-2026-diving</identifier>
    <location>
        <url>https://aclanthology.org/2026.acl-short.74/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>898</start>
            <end>906</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Diving into the Decoding Space of Non-Autoregressive Models via Lexically Constrained Search
%A Huang, Chenyang
%A Zaiane, Osmar
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-391-3
%F huang-zaiane-2026-diving
%X Non-autoregressive (NAR) models have been mainly developed to improve decoding efficiency. Lately, they have also shown great potential in controlled text generation tasks. In this work, we investigate the decoding space of NAR models through lexically constrained machine translation tasks, and develop a search-based decoding algorithm named LexMAP, which is comparable to the autoregressive Grid Beam Search (GBS) method. Our analysis reveals several interesting properties of NAR decoding: 1) the NAR-based method does not suffer from the MAP degradation issue as the autoregressive method does; 2) AR beam search exhibits strong positional bias, in which the candidates only diverge at the end of the sequence; 3) NAR search explores a larger portion of the probability space, suggesting that the search algorithm better exploits the model’s potential.
%U https://aclanthology.org/2026.acl-short.74/
%P 898-906

Markdown (Informal)

[Diving into the Decoding Space of Non-Autoregressive Models via Lexically Constrained Search](https://aclanthology.org/2026.acl-short.74/) (Huang & Zaiane, ACL 2026)

• Diving into the Decoding Space of Non-Autoregressive Models via Lexically Constrained Search (Huang & Zaiane, ACL 2026)

ACL

• Chenyang Huang and Osmar Zaiane. 2026. Diving into the Decoding Space of Non-Autoregressive Models via Lexically Constrained Search. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 898–906, San Diego, California, United States. Association for Computational Linguistics.