@inproceedings{charkiewicz-etal-2026-findings,
title = "Findings of the {WMT}26 Terminology Translation Task: The Hard Part is Finding the Terms, Not Using Them",
author = "Charkiewicz, Adrian and
Chen, Pinzhen and
Etchegoyhen, Thierry and
Gete, Harritxu and
Guttmann, Kamil and
Huang, Xu and
Ponce, David and
Nowakowski, Artur and
Odermatt, Frederic and
Oncevay, Arturo and
Zhu, Dawei and
Zouhar, Vilem and
Semenov, Kirill",
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.51/",
pages = "1046--1079",
ISBN = "979-8-89176-474-3",
abstract = "The WMT26 Terminology Translation Task aims to evaluate machine translation in high-stakes, term-heavy domains (technology, finance, medicine). This year, we focus solely on document-level translation and run two tasks: (1) MT with explicit document-level dictionaries, (2) MT with bitext samples that contain the specific terms. Participants are presented with the texts in three translation directions, two of which feature mid-to-low-resourced morphologically rich languages: Spanish{\$}{\{}{\textbackslash}rightarrow{\}}{\$}Basque, English{\$}{\{}{\textbackslash}rightarrow{\}}{\$}Polish, and Traditional Chinese{\$}{\{}{\textbackslash}rightarrow{\}}{\$}English. This year, the main metrics were multiple variants of overall translation quality and terminology success rate; in line with previous shared tasks, we also compared systems with no terminology, proper dictionaries, and random dictionaries to causally analyze terminology utility. 17 teams participated in our task, submitting 21 systems to Track 1 and 18 systems to Track 2. The results show that document-level translation with explicit dictionaries is close to saturation, and the best systems nearly reach the reference texts. In contrast, for translation with sample bitexts, the spread of the systems is bigger, and the best scores are lower, highlighting the need to concentrate on terminology extraction rather than its use from an explicit source. We also evaluate the grammaticality of the generated texts in Basque and Polish and observe slight trends toward using dictionary forms of the terms and assigning the most frequent grammemes."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="charkiewicz-etal-2026-findings">
<titleInfo>
<title>Findings of the WMT26 Terminology Translation Task: The Hard Part is Finding the Terms, Not Using Them</title>
</titleInfo>
<name type="personal">
<namePart type="given">Adrian</namePart>
<namePart type="family">Charkiewicz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pinzhen</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thierry</namePart>
<namePart type="family">Etchegoyhen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Harritxu</namePart>
<namePart type="family">Gete</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kamil</namePart>
<namePart type="family">Guttmann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xu</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Ponce</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artur</namePart>
<namePart type="family">Nowakowski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Frederic</namePart>
<namePart type="family">Odermatt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arturo</namePart>
<namePart type="family">Oncevay</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dawei</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vilem</namePart>
<namePart type="family">Zouhar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kirill</namePart>
<namePart type="family">Semenov</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>The WMT26 Terminology Translation Task aims to evaluate machine translation in high-stakes, term-heavy domains (technology, finance, medicine). This year, we focus solely on document-level translation and run two tasks: (1) MT with explicit document-level dictionaries, (2) MT with bitext samples that contain the specific terms. Participants are presented with the texts in three translation directions, two of which feature mid-to-low-resourced morphologically rich languages: Spanish${\textbackslashrightarrow}$Basque, English${\textbackslashrightarrow}$Polish, and Traditional Chinese${\textbackslashrightarrow}$English. This year, the main metrics were multiple variants of overall translation quality and terminology success rate; in line with previous shared tasks, we also compared systems with no terminology, proper dictionaries, and random dictionaries to causally analyze terminology utility. 17 teams participated in our task, submitting 21 systems to Track 1 and 18 systems to Track 2. The results show that document-level translation with explicit dictionaries is close to saturation, and the best systems nearly reach the reference texts. In contrast, for translation with sample bitexts, the spread of the systems is bigger, and the best scores are lower, highlighting the need to concentrate on terminology extraction rather than its use from an explicit source. We also evaluate the grammaticality of the generated texts in Basque and Polish and observe slight trends toward using dictionary forms of the terms and assigning the most frequent grammemes.</abstract>
<identifier type="citekey">charkiewicz-etal-2026-findings</identifier>
<location>
<url>https://aclanthology.org/2026.wmt-main.51/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1046</start>
<end>1079</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Findings of the WMT26 Terminology Translation Task: The Hard Part is Finding the Terms, Not Using Them
%A Charkiewicz, Adrian
%A Chen, Pinzhen
%A Etchegoyhen, Thierry
%A Gete, Harritxu
%A Guttmann, Kamil
%A Huang, Xu
%A Ponce, David
%A Nowakowski, Artur
%A Odermatt, Frederic
%A Oncevay, Arturo
%A Zhu, Dawei
%A Zouhar, Vilem
%A Semenov, Kirill
%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 charkiewicz-etal-2026-findings
%X The WMT26 Terminology Translation Task aims to evaluate machine translation in high-stakes, term-heavy domains (technology, finance, medicine). This year, we focus solely on document-level translation and run two tasks: (1) MT with explicit document-level dictionaries, (2) MT with bitext samples that contain the specific terms. Participants are presented with the texts in three translation directions, two of which feature mid-to-low-resourced morphologically rich languages: Spanish${\textbackslashrightarrow}$Basque, English${\textbackslashrightarrow}$Polish, and Traditional Chinese${\textbackslashrightarrow}$English. This year, the main metrics were multiple variants of overall translation quality and terminology success rate; in line with previous shared tasks, we also compared systems with no terminology, proper dictionaries, and random dictionaries to causally analyze terminology utility. 17 teams participated in our task, submitting 21 systems to Track 1 and 18 systems to Track 2. The results show that document-level translation with explicit dictionaries is close to saturation, and the best systems nearly reach the reference texts. In contrast, for translation with sample bitexts, the spread of the systems is bigger, and the best scores are lower, highlighting the need to concentrate on terminology extraction rather than its use from an explicit source. We also evaluate the grammaticality of the generated texts in Basque and Polish and observe slight trends toward using dictionary forms of the terms and assigning the most frequent grammemes.
%U https://aclanthology.org/2026.wmt-main.51/
%P 1046-1079Markdown (Informal)
[Findings of the WMT26 Terminology Translation Task: The Hard Part is Finding the Terms, Not Using Them](https://aclanthology.org/2026.wmt-main.51/) (Charkiewicz et al., WMT 2026)
ACL
- Adrian Charkiewicz, Pinzhen Chen, Thierry Etchegoyhen, Harritxu Gete, Kamil Guttmann, Xu Huang, David Ponce, Artur Nowakowski, Frederic Odermatt, Arturo Oncevay, Dawei Zhu, Vilem Zouhar, and Kirill Semenov. 2026. Findings of the WMT26 Terminology Translation Task: The Hard Part is Finding the Terms, Not Using Them. In Proceedings of the Eleventh Conference on Machine Translation, pages 1046–1079, Budapest, Hungary. Association for Computational Linguistics.