Year: 2026
Type: conference-paper
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 Track 1 (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 Track 2 (in-domain seed bitexts) we compare presenting them as translation history, extracting a dictionary and reusing the Track 1 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 Track 1 submission ties for the highest COMET-22 score and obtains the best XCOMET-XXL and MetricX-24 XL scores, and our primary Track 2 submission achieves the highest term success.
Authors: Aleš Manuel Papáček, Tomáš Sourada
Institution: Charles University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
Funder: European Commission
OpenAlex ID: W7224544707