IRB-MT at WMT25 Terminology Translation Task: Metric-guided Multi-agent Approach

Grubišić, Ivan; Korenčić, Damir (2025) IRB-MT at WMT25 Terminology Translation Task: Metric-guided Multi-agent Approach. In: Haddow, Barry; Kocmi, Tom; Koehn, Philipp; Monz, Christof, (eds.) Proceedings of the Tenth Conference on Machine Translation. Suzhou, Kina, Association for Computational Linguistics (ACL), pp. 1302-1334 .

	PDF - Published Version - article Available under License Creative Commons Attribution. Download (352kB)
	PDF - Published Version - poster Available under License Creative Commons Attribution. Download (33kB)

Abstract

Terminology-aware machine translation (MT) is needed in case of specialized domains such as science and law. Large Language Models (LLMs) have raised the level of state-of-art performance on the task of MT, but the problem is not completely solved, especially for use-cases requiring precise terminology translations. We participate in the WMT25 Terminology Translation Task with an LLM-based multi-agent system coupled with a custom terminology-aware translation quality metric for the selection of the final translation. We use a number of smaller open-weights LLMs embedded in an agentic “translation revision” workflow, and we do not rely on data- and compute-intensive fine-tuning of models. Our evaluations show that the system achieves very good results in terms of both MetricX-24 and a custom TSR metric designed to measure the adherence to predefined term mappings.

Item Type:	Conference or workshop item published in conference proceedings (UNSPECIFIED)
Uncontrolled Keywords:	machine translation; terminology translation; natural language processing; large language models; multi-agent system
Subjects:	TECHNICAL SCIENCES > Computing > Artificial Intelligence
Divisions:	Division of Electronics
Depositing User:	Ivan Grubišić
Date Deposited:	14 Jan 2026 15:25
URI:	http://fulir.irb.hr/id/eprint/10841
DOI:	10.18653/v1/2025.wmt-1.110

Actions (login required)

View Item