GRAFT: A Graph-based Flow-aware Agentic Framework for Document-level Machine Translation
This addresses the challenge of maintaining consistency and accuracy in document-level machine translation for applications requiring coherent translations across diverse domains.
The paper tackled the problem of capturing discourse-level phenomena in document-level machine translation by proposing GRAFT, a graph-based framework using LLM agents, which achieved average improvements of 2.8 d BLEU on TED test sets and 2.3 d BLEU for domain-specific translation.
Document level Machine Translation (DocMT) approaches often struggle with effectively capturing discourse level phenomena. Existing approaches rely on heuristic rules to segment documents into discourse units, which rarely align with the true discourse structure required for accurate translation. Otherwise, they fail to maintain consistency throughout the document during translation. To address these challenges, we propose Graph Augmented Agentic Framework for Document Level Translation (GRAFT), a novel graph based DocMT system that leverages Large Language Model (LLM) agents for document translation. Our approach integrates segmentation, directed acyclic graph (DAG) based dependency modelling, and discourse aware translation into a cohesive framework. Experiments conducted across eight translation directions and six diverse domains demonstrate that GRAFT achieves significant performance gains over state of the art DocMT systems. Specifically, GRAFT delivers an average improvement of 2.8 d BLEU on the TED test sets from IWSLT2017 over strong baselines and 2.3 d BLEU for domain specific translation from English to Chinese. Moreover, our analyses highlight the consistent ability of GRAFT to address discourse level phenomena, yielding coherent and contextually accurate translations.