DelTA: An Online Document-Level Translation Agent Based on Multi-Level Memory

2024-10-10

arXiv.org

AI Summary

To address insufficient consistency and accuracy in document-level machine translation, this paper proposes DelTA, an intelligent agent designed for online document translation. Its core is a novel multi-granularity dynamic memory architecture that integrates a proper noun lexicon, bilingual summaries, and short- and long-term memory modules, enabling large language models (LLMs) to perform incremental sentence-level translation that is consistent, accurate, and complete. DelTA is the first framework to unify translation agents with query-driven summarization capabilities, featuring an LLM-guided memory retrieval and update module alongside a hierarchical memory storage mechanism. Evaluated on two benchmark datasets, DelTA achieves average improvements of +4.58 points in consistency score and +3.16 points in COMET score, significantly enhancing pronoun resolution and context-dependent translation. The code and datasets are publicly available.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Multiagent Systems: Multiagent Learning

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Large language models (LLMs) have achieved reasonable quality improvements in machine translation (MT). However, most current research on MT-LLMs still faces significant challenges in maintaining translation consistency and accuracy when processing entire documents. In this paper, we introduce DelTA, a Document-levEL Translation Agent designed to overcome these limitations. DelTA features a multi-level memory structure that stores information across various granularities and spans, including Proper Noun Records, Bilingual Summary, Long-Term Memory, and Short-Term Memory, which are continuously retrieved and updated by auxiliary LLM-based components. Experimental results indicate that DelTA significantly outperforms strong baselines in terms of translation consistency and quality across four open/closed-source LLMs and two representative document translation datasets, achieving an increase in consistency scores by up to 4.58 percentage points and in COMET scores by up to 3.16 points on average. DelTA employs a sentence-by-sentence translation strategy, ensuring no sentence omissions and offering a memory-efficient solution compared to the mainstream method. Furthermore, DelTA improves pronoun and context-dependent translation accuracy, and the summary component of the agent also shows promise as a tool for query-based summarization tasks. The code and data of our approach are released at https://github.com/YutongWang1216/DocMTAgent.

Problem

• Improves document-level translation consistency and accuracy
• Addresses challenges in maintaining translation quality across entire documents
• Enhances pronoun and context-dependent translation accuracy

Innovation

• Multi-level memory structure for translation consistency
• Sentence-by-sentence strategy prevents omissions
• Auxiliary LLM components enhance context accuracy