Structured Document Translation via Format Reinforcement Learning
This addresses the challenge of document-level structured text translation for applications like software documentation, though it is incremental as it builds on existing reinforcement learning methods.
The paper tackles the problem of translating structured documents like XML or HTML by proposing Format Reinforcement Learning (FormatRL), which uses novel structure-aware rewards to improve both structural and translation quality, achieving improvements across six metrics on a software-documentation benchmark.
Recent works on structured text translation remain limited to the sentence level, as they struggle to effectively handle the complex document-level XML or HTML structures. To address this, we propose \textbf{Format Reinforcement Learning (FormatRL)}, which employs Group Relative Policy Optimization on top of a supervised fine-tuning model to directly optimize novel structure-aware rewards: 1) TreeSim, which measures structural similarity between predicted and reference XML trees and 2) Node-chrF, which measures translation quality at the level of XML nodes. Additionally, we apply StrucAUC, a fine-grained metric distinguishing between minor errors and major structural failures. Experiments on the SAP software-documentation benchmark demonstrate improvements across six metrics and an analysis further shows how different reward functions contribute to improvements in both structural and translation quality.