Context-aware and Style-related Incremental Decoding framework for Discourse-Level Literary Translation

TL;DR

This paper presents a novel context-aware incremental decoding framework for Chinese-English literary translation, enhancing coherence and style preservation by leveraging a specialized Llama2 model with continual pre-training and fine-tuning.

Contribution

The work introduces a new incremental decoding approach combined with a specialized Chinese-Llama2 model for improved discourse-level literary translation.

Findings

Significant BLEU score improvements at sentence and document levels

Effective preservation of literary style and coherence

Demonstrated capability to handle long-range dependencies

Abstract

This report outlines our approach for the WMT24 Discourse-Level Literary Translation Task, focusing on the Chinese-English language pair in the Constrained Track. Translating literary texts poses significant challenges due to the nuanced meanings, idiomatic expressions, and intricate narrative structures inherent in such works. To address these challenges, we leveraged the Chinese-Llama2 model, specifically enhanced for this task through a combination of Continual Pre-training (CPT) and Supervised Fine-Tuning (SFT). Our methodology includes a novel Incremental Decoding framework, which ensures that each sentence is translated with consideration of its broader context, maintaining coherence and consistency throughout the text. This approach allows the model to capture long-range dependencies and stylistic elements, producing translations that faithfully preserve the original literary quality. Our experiments demonstrate significant improvements in both sentence-level and document-level BLEU scores, underscoring the effectiveness of our proposed framework in addressing the complexities of document-level literary translation.