Mitigating Stylistic Biases of Machine Translation Systems via Monolingual Corpora Only

TL;DR

This paper introduces Babel, a novel framework that enhances stylistic fidelity in neural machine translation using only monolingual data, employing style detection and diffusion-based style application to improve style preservation without modifying existing systems.

Contribution

Babel is a new post-processing framework that improves stylistic accuracy in NMT using monolingual corpora, avoiding the need for parallel stylistic data or system modifications.

Findings

Achieves 88.21% precision in detecting stylistic inconsistencies

Improves stylistic preservation by 150%

Maintains a semantic similarity score of 0.92

Abstract

The advent of neural machine translation (NMT) has revolutionized cross-lingual communication, yet preserving stylistic nuances remains a significant challenge. While existing approaches often require parallel corpora for style preservation, we introduce Babel, a novel framework that enhances stylistic fidelity in NMT using only monolingual corpora. Babel employs two key components: (1) a style detector based on contextual embeddings that identifies stylistic disparities between source and target texts, and (2) a diffusion-based style applicator that rectifies stylistic inconsistencies while maintaining semantic integrity. Our framework integrates with existing NMT systems as a post-processing module, enabling style-aware translation without requiring architectural modifications or parallel stylistic data. Extensive experiments on five diverse domains (law, literature, scientific writing, medicine, and educational content) demonstrate Babel's effectiveness: it identifies stylistic inconsistencies with 88.21% precision and improves stylistic preservation by 150% while maintaining a high semantic similarity score of 0.92. Human evaluation confirms that translations refined by Babel better preserve source text style while maintaining fluency and adequacy.