@inproceedings{papacek-sourada-2026-cuni,
title = "{CUNI}-{{\'U}FAL} at the {WMT}26 Terminology Translation Task",
author = "Pap{\'a}{\v{c}}ek, Ale{\v{s}} Manuel and
Sourada, Tom{\'a}{\v{s}}",
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.120/",
pages = "1849--1859",
ISBN = "979-8-89176-474-3",
abstract = "This paper describes the CUNI-{\'U}FAL submission to the WMT26 Terminology Translation Task. We present a model-agnostic pipeline that selects document-relevant terminology for a general-purpose LLM, instantiated with GPT-5.6-Sol for both tracks. For Track1 (explicit dictionary) we compare glossary placement, inline annotation, dictionary filtering, and terminology-aware revision: filtering the dictionary per document improves judged coverage and, slightly, translation quality. Inline term insertion buys a small coverage gain at the cost of fluency. For Track2 (in-domain seed bitexts) we compare presenting them as translation history, extracting a dictionary and reusing the Track1 pipeline, and their hybrid. Lacking development references, we evaluated with surface- and lemma-based term matching plus an LLM judge. Among submitted systems in the official evaluation, our Track1 submission ties for the highest COMET-22 score and obtains the best XCOMET-XXL and MetricX-24 XL scores, and our primary Track2 submission achieves the highest term success."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="papacek-sourada-2026-cuni">
<titleInfo>
<title>CUNI-ÚFAL at the WMT26 Terminology Translation Task</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aleš</namePart>
<namePart type="given">Manuel</namePart>
<namePart type="family">Papáček</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tomáš</namePart>
<namePart type="family">Sourada</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 the CUNI-ÚFAL submission to the WMT26 Terminology Translation Task. We present a model-agnostic pipeline that selects document-relevant terminology for a general-purpose LLM, instantiated with GPT-5.6-Sol for both tracks. For Track1 (explicit dictionary) we compare glossary placement, inline annotation, dictionary filtering, and terminology-aware revision: filtering the dictionary per document improves judged coverage and, slightly, translation quality. Inline term insertion buys a small coverage gain at the cost of fluency. For Track2 (in-domain seed bitexts) we compare presenting them as translation history, extracting a dictionary and reusing the Track1 pipeline, and their hybrid. Lacking development references, we evaluated with surface- and lemma-based term matching plus an LLM judge. Among submitted systems in the official evaluation, our Track1 submission ties for the highest COMET-22 score and obtains the best XCOMET-XXL and MetricX-24 XL scores, and our primary Track2 submission achieves the highest term success.</abstract>
<identifier type="citekey">papacek-sourada-2026-cuni</identifier>
<location>
<url>https://aclanthology.org/2026.wmt-main.120/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1849</start>
<end>1859</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CUNI-ÚFAL at the WMT26 Terminology Translation Task
%A Papáček, Aleš Manuel
%A Sourada, Tomáš
%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 papacek-sourada-2026-cuni
%X This paper describes the CUNI-ÚFAL submission to the WMT26 Terminology Translation Task. We present a model-agnostic pipeline that selects document-relevant terminology for a general-purpose LLM, instantiated with GPT-5.6-Sol for both tracks. For Track1 (explicit dictionary) we compare glossary placement, inline annotation, dictionary filtering, and terminology-aware revision: filtering the dictionary per document improves judged coverage and, slightly, translation quality. Inline term insertion buys a small coverage gain at the cost of fluency. For Track2 (in-domain seed bitexts) we compare presenting them as translation history, extracting a dictionary and reusing the Track1 pipeline, and their hybrid. Lacking development references, we evaluated with surface- and lemma-based term matching plus an LLM judge. Among submitted systems in the official evaluation, our Track1 submission ties for the highest COMET-22 score and obtains the best XCOMET-XXL and MetricX-24 XL scores, and our primary Track2 submission achieves the highest term success.
%U https://aclanthology.org/2026.wmt-main.120/
%P 1849-1859
Markdown (Informal)
[CUNI-ÚFAL at the WMT26 Terminology Translation Task](https://aclanthology.org/2026.wmt-main.120/) (Papáček & Sourada, WMT 2026)
ACL