@inproceedings{su-etal-2026-jiutian,
title = "Jiutian-{MT}-8{B} at {WMT}26: Faithfulness-First Training for {C}hinese-{S}outheast {A}sian Translation",
author = "Su, Meng and
Liu, Jian and
Fang, Youxuan",
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.181/",
pages = "2374--2378",
ISBN = "979-8-89176-474-3",
abstract = "This paper describes Jiutian-MT-8B, our submission to the WMT 2026 Chinese-Southeast Asian Multilingual Machine Translation shared task. The system adapts the 8-billion-parameter Jiutian-8B-Instruct checkpoint to fourteen translation directions between Chinese and Thai, Vietnamese, Lao, Burmese, Khmer, Indonesian, and Malay. We formulate translation as a tightly controlled instruction-following task and use a four-stage full-parameter training recipe: (i) supervised multilingual adaptation on released parallel data, (ii) faithfulness first confidence-gated multi-pair preference alignment (F2MTA), (iii) direction-coupled, evidence-filtered self-training from task monolingual data, and (iv) a final clean-data calibration stage. The preference stage uses a partial order: a candidate must gain source adequacy without materially sacrificing target fluency, rather than merely win a scalarized quality score. The recipe directly updates all Jiutian-8B parameters with sharded bf16 training while retaining authentic data anchors to limit synthetic-data drift. The official test set is blind at the time of writing; numerical main results are therefore deliberately unreported. We nevertheless specify the data scope, translation interface, training protocol, and evaluation plan needed to complete the report once official scores are returned."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="su-etal-2026-jiutian">
<titleInfo>
<title>Jiutian-MT-8B at WMT26: Faithfulness-First Training for Chinese-Southeast Asian Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Meng</namePart>
<namePart type="family">Su</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jian</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Youxuan</namePart>
<namePart type="family">Fang</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>This paper describes Jiutian-MT-8B, our submission to the WMT 2026 Chinese-Southeast Asian Multilingual Machine Translation shared task. The system adapts the 8-billion-parameter Jiutian-8B-Instruct checkpoint to fourteen translation directions between Chinese and Thai, Vietnamese, Lao, Burmese, Khmer, Indonesian, and Malay. We formulate translation as a tightly controlled instruction-following task and use a four-stage full-parameter training recipe: (i) supervised multilingual adaptation on released parallel data, (ii) faithfulness first confidence-gated multi-pair preference alignment (F2MTA), (iii) direction-coupled, evidence-filtered self-training from task monolingual data, and (iv) a final clean-data calibration stage. The preference stage uses a partial order: a candidate must gain source adequacy without materially sacrificing target fluency, rather than merely win a scalarized quality score. The recipe directly updates all Jiutian-8B parameters with sharded bf16 training while retaining authentic data anchors to limit synthetic-data drift. The official test set is blind at the time of writing; numerical main results are therefore deliberately unreported. We nevertheless specify the data scope, translation interface, training protocol, and evaluation plan needed to complete the report once official scores are returned.</abstract>
<identifier type="citekey">su-etal-2026-jiutian</identifier>
<location>
<url>https://aclanthology.org/2026.wmt-main.181/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>2374</start>
<end>2378</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Jiutian-MT-8B at WMT26: Faithfulness-First Training for Chinese-Southeast Asian Translation
%A Su, Meng
%A Liu, Jian
%A Fang, Youxuan
%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 su-etal-2026-jiutian
%X This paper describes Jiutian-MT-8B, our submission to the WMT 2026 Chinese-Southeast Asian Multilingual Machine Translation shared task. The system adapts the 8-billion-parameter Jiutian-8B-Instruct checkpoint to fourteen translation directions between Chinese and Thai, Vietnamese, Lao, Burmese, Khmer, Indonesian, and Malay. We formulate translation as a tightly controlled instruction-following task and use a four-stage full-parameter training recipe: (i) supervised multilingual adaptation on released parallel data, (ii) faithfulness first confidence-gated multi-pair preference alignment (F2MTA), (iii) direction-coupled, evidence-filtered self-training from task monolingual data, and (iv) a final clean-data calibration stage. The preference stage uses a partial order: a candidate must gain source adequacy without materially sacrificing target fluency, rather than merely win a scalarized quality score. The recipe directly updates all Jiutian-8B parameters with sharded bf16 training while retaining authentic data anchors to limit synthetic-data drift. The official test set is blind at the time of writing; numerical main results are therefore deliberately unreported. We nevertheless specify the data scope, translation interface, training protocol, and evaluation plan needed to complete the report once official scores are returned.
%U https://aclanthology.org/2026.wmt-main.181/
%P 2374-2378
Markdown (Informal)
[Jiutian-MT-8B at WMT26: Faithfulness-First Training for Chinese-Southeast Asian Translation](https://aclanthology.org/2026.wmt-main.181/) (Su et al., WMT 2026)
ACL