@inproceedings{matsuda-etal-2026-rfmt,
title = "{RFMT} at {WMT} 2026 General Translation Task",
author = "Matsuda, Ryosuke and
Kudo, Keito and
Fujii, Ryo and
Ito, Takumi and
Morishita, Makoto and
Suzuki, Jun",
editor = "Haddow, Barry and
Kocmi, Tom and
Koehn, Philipp and
Monz, Christof",
booktitle = "Proceedings of the Eleventh Conference on Machine Translation",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.wmt-main.73/",
pages = "1344--1371",
ISBN = "979-8-89176-474-3",
abstract = "We participated in the constrained (open-weight) track of the WMT 2026 General MT Task for the English-to-Japanese and Simplified Chinese-to-Japanese directions. Our system builds on Marco-MT-Algharb, a strong open-weight translation model from the previous WMT General MT task. We further fine-tuned this model to generate more natural Japanese translations. To this end, we constructed a Context-Aware JApanese Linguistic Acceptability dataset (CAJALA). During annotation, human annotators were shown multiple paraphrased variants of a segment within its document-level context and asked to select the most natural one. We then back-translated the CAJALA dataset to create parallel data for Direct Preference Optimization (DPO). We also trained the model to perform fill-in-the-middle (FIM) translation, in which the model reconstructs target-side segments from their surrounding document. At inference, we use Minimum Bayes Risk (MBR) decoding, followed by selective FIM post-editing of low-quality segments."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="matsuda-etal-2026-rfmt">
<titleInfo>
<title>RFMT at WMT 2026 General Translation Task</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ryosuke</namePart>
<namePart type="family">Matsuda</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Keito</namePart>
<namePart type="family">Kudo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryo</namePart>
<namePart type="family">Fujii</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Takumi</namePart>
<namePart type="family">Ito</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Makoto</namePart>
<namePart type="family">Morishita</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</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 Eleventh Conference on Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Barry</namePart>
<namePart type="family">Haddow</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tom</namePart>
<namePart type="family">Kocmi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Philipp</namePart>
<namePart type="family">Koehn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christof</namePart>
<namePart type="family">Monz</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">979-8-89176-474-3</identifier>
</relatedItem>
<abstract>We participated in the constrained (open-weight) track of the WMT 2026 General MT Task for the English-to-Japanese and Simplified Chinese-to-Japanese directions. Our system builds on Marco-MT-Algharb, a strong open-weight translation model from the previous WMT General MT task. We further fine-tuned this model to generate more natural Japanese translations. To this end, we constructed a Context-Aware JApanese Linguistic Acceptability dataset (CAJALA). During annotation, human annotators were shown multiple paraphrased variants of a segment within its document-level context and asked to select the most natural one. We then back-translated the CAJALA dataset to create parallel data for Direct Preference Optimization (DPO). We also trained the model to perform fill-in-the-middle (FIM) translation, in which the model reconstructs target-side segments from their surrounding document. At inference, we use Minimum Bayes Risk (MBR) decoding, followed by selective FIM post-editing of low-quality segments.</abstract>
<identifier type="citekey">matsuda-etal-2026-rfmt</identifier>
<location>
<url>https://aclanthology.org/2026.wmt-main.73/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1344</start>
<end>1371</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T RFMT at WMT 2026 General Translation Task
%A Matsuda, Ryosuke
%A Kudo, Keito
%A Fujii, Ryo
%A Ito, Takumi
%A Morishita, Makoto
%A Suzuki, Jun
%Y Haddow, Barry
%Y Kocmi, Tom
%Y Koehn, Philipp
%Y Monz, Christof
%S Proceedings of the Eleventh Conference on Machine Translation
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-474-3
%F matsuda-etal-2026-rfmt
%X We participated in the constrained (open-weight) track of the WMT 2026 General MT Task for the English-to-Japanese and Simplified Chinese-to-Japanese directions. Our system builds on Marco-MT-Algharb, a strong open-weight translation model from the previous WMT General MT task. We further fine-tuned this model to generate more natural Japanese translations. To this end, we constructed a Context-Aware JApanese Linguistic Acceptability dataset (CAJALA). During annotation, human annotators were shown multiple paraphrased variants of a segment within its document-level context and asked to select the most natural one. We then back-translated the CAJALA dataset to create parallel data for Direct Preference Optimization (DPO). We also trained the model to perform fill-in-the-middle (FIM) translation, in which the model reconstructs target-side segments from their surrounding document. At inference, we use Minimum Bayes Risk (MBR) decoding, followed by selective FIM post-editing of low-quality segments.
%U https://aclanthology.org/2026.wmt-main.73/
%P 1344-1371
Markdown (Informal)
[RFMT at WMT 2026 General Translation Task](https://aclanthology.org/2026.wmt-main.73/) (Matsuda et al., WMT 2026)
ACL
- Ryosuke Matsuda, Keito Kudo, Ryo Fujii, Takumi Ito, Makoto Morishita, and Jun Suzuki. 2026. RFMT at WMT 2026 General Translation Task. In Proceedings of the Eleventh Conference on Machine Translation, pages 1344–1371, Budapest, Hungary. Association for Computational Linguistics.