FeedTrans: Enhance Machine Translation via Feedback

Zengkui Sun; Jiali Zeng; Jiaan Wang; Fandong Meng; Xinyan Guan; Yufeng Chen (陈钰枫); Jinan Xu (徐金安); Wenjuan Han; Jie Zhou

doi:10.1162/tacl.a.784

FeedTrans: Enhance Machine Translation via Feedback

Zengkui Sun, Jiali Zeng, Jiaan Wang, Fandong Meng, Xinyan Guan, Yufeng Chen, Jinan Xu, Wenjuan Han, Jie Zhou

Correct Metadata for

Use this form to create a GitHub issue with structured data describing the correction. You will need a GitHub account. Once you create that issue, the correction will be reviewed by a staff member.

⚠️ Mobile Users: Submitting this form to create a new issue will only work with github.com, not the GitHub Mobile app.

Important: The Anthology treat PDFs as authoritative. Please use this form only to correct data that is out of line with the PDF. See our corrections guidelines if you need to change the PDF.

Title Adjust the title. Retain tags such as <fixed-case>.

Authors Adjust author names and order to match the PDF.

Abstract Correct abstract if needed. Retain XML formatting tags such as <tex-math>. You may use ... for bold, ... for italic, ... for underline, <sc>...</sc> for small-caps, <tt>...<tt> for typewriter text, <url>...</url> for URLs, <a href=...> for hyperlinks, and <par/> for paragraph breaks.

Verification against PDF Ensure that the new title/authors match the snapshot below. (If there is no snapshot or it is too small, consult the PDF.)

Authors concatenated from the text boxes above:

ALL author names match the snapshot above—including middle initials, hyphens, and accents.

Abstract

Large reasoning models (LRMs) have shown exceptional performance in complex tasks such as mathematics and coding. In the field of machine translation (MT), reinforcement learning (RL) has been utilized to enhance the quality of translations. However, traditional RL approaches rely heavily on the base model’s inherent translation capabilities, which may falter when dealing with terminology translations and domain-specific expressions without sufficient guidance. In this paper, we introduce FeedTrans (Feedback-driven Translation), which employs a FeedRollout mechanism to incorporate feedback as guidance, enabling the production of high-quality translations and expanding the search space for improved translation outcomes. Extensive experiments across six translation tasks validate the effectiveness of our approach. To the best of our knowledge, this is the first attempt at utilizing feedback in MT-oriented RL. https://github.com/Acerkoo/FeedTrans

Anthology ID:: 2026.tacl-1.86
Volume:: Transactions of the Association for Computational Linguistics, Volume 14
Month:
Year:: 2026
Address:: Cambridge, MA
Venue:: TACL
SIG:
Publisher:: MIT Press
Note:
Pages:: 1899–1916
Language:
URL:: https://aclanthology.org/2026.tacl-1.86/
DOI:: 10.1162/tacl.a.784
Bibkey:
Cite (ACL):: Zengkui Sun, Jiali Zeng, Jiaan Wang, Fandong Meng, Xinyan Guan, Yufeng Chen, Jinan Xu, Wenjuan Han, and Jie Zhou. 2026. FeedTrans: Enhance Machine Translation via Feedback. Transactions of the Association for Computational Linguistics, 14:1899–1916.
Cite (Informal):: FeedTrans: Enhance Machine Translation via Feedback (Sun et al., TACL 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.tacl-1.86.pdf

PDF Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@article{sun-etal-2026-feedtrans,
    title = "{F}eed{T}rans: Enhance Machine Translation via Feedback",
    author = "Sun, Zengkui  and
      Zeng, Jiali  and
      Wang, Jiaan  and
      Meng, Fandong  and
      Guan, Xinyan  and
      Chen, Yufeng  and
      Xu, Jinan  and
      Han, Wenjuan  and
      Zhou, Jie",
    journal = "Transactions of the Association for Computational Linguistics",
    volume = "14",
    year = "2026",
    address = "Cambridge, MA",
    publisher = "MIT Press",
    url = "https://aclanthology.org/2026.tacl-1.86/",
    doi = "10.1162/tacl.a.784",
    pages = "1899--1916",
    abstract = "Large reasoning models (LRMs) have shown exceptional performance in complex tasks such as mathematics and coding. In the field of machine translation (MT), reinforcement learning (RL) has been utilized to enhance the quality of translations. However, traditional RL approaches rely heavily on the base model{'}s inherent translation capabilities, which may falter when dealing with terminology translations and domain-specific expressions without sufficient guidance. In this paper, we introduce FeedTrans (Feedback-driven Translation), which employs a FeedRollout mechanism to incorporate feedback as guidance, enabling the production of high-quality translations and expanding the search space for improved translation outcomes. Extensive experiments across six translation tasks validate the effectiveness of our approach. To the best of our knowledge, this is the first attempt at utilizing feedback in MT-oriented RL. https://github.com/Acerkoo/FeedTrans"
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sun-etal-2026-feedtrans">
    <titleInfo>
        <title>FeedTrans: Enhance Machine Translation via Feedback</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Zengkui</namePart>
        <namePart type="family">Sun</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiali</namePart>
        <namePart type="family">Zeng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiaan</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Fandong</namePart>
        <namePart type="family">Meng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xinyan</namePart>
        <namePart type="family">Guan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yufeng</namePart>
        <namePart type="family">Chen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jinan</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Wenjuan</namePart>
        <namePart type="family">Han</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jie</namePart>
        <namePart type="family">Zhou</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>Large reasoning models (LRMs) have shown exceptional performance in complex tasks such as mathematics and coding. In the field of machine translation (MT), reinforcement learning (RL) has been utilized to enhance the quality of translations. However, traditional RL approaches rely heavily on the base model’s inherent translation capabilities, which may falter when dealing with terminology translations and domain-specific expressions without sufficient guidance. In this paper, we introduce FeedTrans (Feedback-driven Translation), which employs a FeedRollout mechanism to incorporate feedback as guidance, enabling the production of high-quality translations and expanding the search space for improved translation outcomes. Extensive experiments across six translation tasks validate the effectiveness of our approach. To the best of our knowledge, this is the first attempt at utilizing feedback in MT-oriented RL. https://github.com/Acerkoo/FeedTrans</abstract>
    <identifier type="citekey">sun-etal-2026-feedtrans</identifier>
    <identifier type="doi">10.1162/tacl.a.784</identifier>
    <location>
        <url>https://aclanthology.org/2026.tacl-1.86/</url>
    </location>
    <part>
        <date>2026</date>
        <detail type="volume"><number>14</number></detail>
        <extent unit="page">
            <start>1899</start>
            <end>1916</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Journal Article
%T FeedTrans: Enhance Machine Translation via Feedback
%A Sun, Zengkui
%A Zeng, Jiali
%A Wang, Jiaan
%A Meng, Fandong
%A Guan, Xinyan
%A Chen, Yufeng
%A Xu, Jinan
%A Han, Wenjuan
%A Zhou, Jie
%J Transactions of the Association for Computational Linguistics
%D 2026
%V 14
%I MIT Press
%C Cambridge, MA
%F sun-etal-2026-feedtrans
%X Large reasoning models (LRMs) have shown exceptional performance in complex tasks such as mathematics and coding. In the field of machine translation (MT), reinforcement learning (RL) has been utilized to enhance the quality of translations. However, traditional RL approaches rely heavily on the base model’s inherent translation capabilities, which may falter when dealing with terminology translations and domain-specific expressions without sufficient guidance. In this paper, we introduce FeedTrans (Feedback-driven Translation), which employs a FeedRollout mechanism to incorporate feedback as guidance, enabling the production of high-quality translations and expanding the search space for improved translation outcomes. Extensive experiments across six translation tasks validate the effectiveness of our approach. To the best of our knowledge, this is the first attempt at utilizing feedback in MT-oriented RL. https://github.com/Acerkoo/FeedTrans
%R 10.1162/tacl.a.784
%U https://aclanthology.org/2026.tacl-1.86/
%U https://doi.org/10.1162/tacl.a.784
%P 1899-1916

Download as File

Markdown (Informal)

[FeedTrans: Enhance Machine Translation via Feedback](https://aclanthology.org/2026.tacl-1.86/) (Sun et al., TACL 2026)

FeedTrans: Enhance Machine Translation via Feedback (Sun et al., TACL 2026)

ACL

Zengkui Sun, Jiali Zeng, Jiaan Wang, Fandong Meng, Xinyan Guan, Yufeng Chen, Jinan Xu, Wenjuan Han, and Jie Zhou. 2026. FeedTrans: Enhance Machine Translation via Feedback. Transactions of the Association for Computational Linguistics, 14:1899–1916.