TAT-R1: Terminology-Aware Translation with Reinforcement Learning and Word Alignment

TL;DR

This paper introduces TAT-R1, a reinforcement learning-based, terminology-aware translation model that improves translation accuracy for key terminology while maintaining overall translation quality.

Contribution

It presents a novel approach combining word alignment and reinforcement learning to enhance terminology translation in deep reasoning large language models.

Findings

Significant improvement in terminology translation accuracy.

Maintains comparable performance on general translation tasks.

Provides insights through ablation studies on training paradigms.

Abstract

Recently, deep reasoning large language models(LLMs) like DeepSeek-R1 have made significant progress in tasks such as mathematics and coding. Inspired by this, several studies have employed reinforcement learning(RL) to enhance models' deep reasoning capabilities and improve machine translation(MT) quality. However, the terminology translation, an essential task in MT, remains unexplored in deep reasoning LLMs. In this paper, we propose \textbf{TAT-R1}, a terminology-aware translation model trained with reinforcement learning and word alignment. Specifically, we first extract the keyword translation pairs using a word alignment model. Then we carefully design three types of rule-based alignment rewards with the extracted alignment relationships. With those alignment rewards, the RL-trained translation model can learn to focus on the accurate translation of key information, including terminology in the source text. Experimental results show the effectiveness of TAT-R1. Our model significantly improves terminology translation accuracy compared to the baseline models while maintaining comparable performance on general translation tasks. In addition, we conduct detailed ablation studies of the DeepSeek-R1-like training paradigm for machine translation and reveal several key findings.