DocGRPO: Adaptive Adequacy and Fluency Preference Optimization for LLM-Based Document-Level Machine Translation

Large language models (LLMs) have shown strong potential for machine translation, yet preserving source information while maintaining coherent discourse remains challenging in document-level machine translation (DocMT). Our preliminary analysis shows that word-alignment-based coverage declines as source documents grow longer, whereas fluency improves and human quality judgments follow a non-monotonic pattern. Coverage alone, however, cannot distinguish genuine omissions from contextually appropriate ellipsis and implicit references, motivating the joint consideration of adequacy and fluency. We propose DocGRPO ($\text{D}$ynamic $\text{O}$bjective $\text{C}$oordination with $\text{G}$roup $\text{R}$elative $\text{P}$olicy $\text{O}$ptimization), a reinforcement-learning framework that combines an alignment-based coverage reward for source-content preservation with a discourse-aware fluency reward. To further alleviate the tension between adequacy and fluency, DocGRPO adapts reward weights to training progress and source-document length while combining group-relative and reference-relative advantages to guide policy updates. Experiments across five translation directions and two model scales show improvements over strong document-level baselines. Human evaluation further confirms better overall translation quality, while ablations support the benefits of complementary rewards and adaptive weighting.