Structured Document Translation via Format Reinforcement Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Song, Haiyue, Eschbach-Dymanus, Johannes, Kaing, Hour, Honda, Sumire, Tanaka, Hideki, Buschbeck, Bianka, Utiyama, Masao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915654327599104
author Song, Haiyue
Eschbach-Dymanus, Johannes
Kaing, Hour
Honda, Sumire
Tanaka, Hideki
Buschbeck, Bianka
Utiyama, Masao
author_facet Song, Haiyue
Eschbach-Dymanus, Johannes
Kaing, Hour
Honda, Sumire
Tanaka, Hideki
Buschbeck, Bianka
Utiyama, Masao
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_05100
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Structured Document Translation via Format Reinforcement Learning
Song, Haiyue
Eschbach-Dymanus, Johannes
Kaing, Hour
Honda, Sumire
Tanaka, Hideki
Buschbeck, Bianka
Utiyama, Masao
Computation and Language
Artificial Intelligence
Machine Learning
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.
title Structured Document Translation via Format Reinforcement Learning
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2512.05100