Enhancing Document-Level Machine Translation via Filtered Synthetic Corpora and Two-Stage LLM Adaptation
This work addresses document-level translation challenges for NLP practitioners, but it is incremental as it builds on existing fine-tuning and data augmentation methods.
The paper tackled the problem of document-level machine translation with large language models by addressing data scarcity and generation issues, resulting in improved translation quality as measured by metrics like BLEU and COMET scores.
In Machine Translation, Large Language Models (LLMs) have generally underperformed compared to conventional encoder-decoder systems and thus see limited adoption. However, LLMs excel at modeling contextual information, making them a natural fit for document-level translation tasks where coherence across sentences is crucial. Despite this potential, document-level MT with LLMs faces two key challenges: (1) the scarcity of large-scale, high-quality document-level parallel data; and (2) the propensity of LLMs to introduce hallucinations and omissions during generation. To address these challenges, we propose a two-stage fine-tuning strategy leveraging LLM-augmented document-level data. First, we augment data by converting summarization data into document-level parallel data using a LLM, and then filter it using multiple metrics, leveraging sacreBLEU, COMET, and LaBSE-based cosine similarity-to improve data quality. Finally, we employ a two-stage fine-tuning strategy: first fine-tuning on the abundant sentence-level MT resources, and then on the filtered document-level corpus.