BridG MT: Enhancing LLMs' Machine Translation Capabilities with Sentence Bridging and Gradual MT
This addresses translation issues for low-resource languages with LLMs, offering an incremental improvement over existing few-shot approaches.
The paper tackles the challenge of low-resource language translation with LLMs by proposing BridG MT, which uses sentence bridging and gradual translation to reduce reliance on external knowledge, resulting in improved performance that surpasses methods using many few-shot examples across four LLMs and seven languages.
Recent Large Language Models (LLMs) have demonstrated impressive translation performance without requiring fine-tuning on additional parallel corpora. However, they still face significant challenges in certain scenarios, particularly when translating low-resource languages. A common approach to address this issue is to provide external knowledge, such as few-shot examples, to assist LLMs in translating specific source sentences. However, this method is fundamentally limited by the quality or quantity of relevant sources, which cannot always be guaranteed. To reduce LLMs' reliance on external sources, we propose BridG MT, a method that combines Sentence Bridging, which generates a sequence of sentences as a bridge that gradually transition from easy-to-translate to more difficult, and Gradual MT, which sequentially translates these sentences using earlier translations as few-shot examples for subsequent ones. Experiments conducted on four LLMs across seven languages demonstrate that our method effectively enhances translation performance, even outperforming translation methods that rely on a large number of few-shot examples.