@inproceedings{shen-etal-2026-pegrl,
title = "{PEGRL}: Improving Machine Translation by Post-Editing Guided Reinforcement Learning",
author = "Shen, Yunzhi and
Zhou, Hao and
Huang, Xin and
Han, Xue and
Feng, Junlan and
Huang, Shujian",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.851/",
pages = "17225--17242",
ISBN = "979-8-89176-395-1",
abstract = "Reinforcement learning (RL) has shown strong promise for LLM-based machine translation, with recent methods such as GRPO demonstrating notable gains; nevertheless, translation-oriented RL remains challenged by high-variance policy gradients induced by Monte Carlo baselines, as well as a large trajectory space that favors global exploration over fine-grained local optimization. We introduce \textbf{PEGRL}, a \textit{two-stage} RL framework that uses post-editing as an auxiliary task to stabilize training and guide overall optimization. At each step, translation outputs are sampled to construct post-editing inputs, enabling lower-variance gradients from the post-editing task to propagate through the entire framework while jointly supporting both global exploration and fine-grained local optimization. A task-specific weighting scheme further emphasizes the post-editing gradient, producing a biased yet more sample-efficient estimator. Experiments on English$\to$Finnish, English$\to$Turkish, and English$\leftrightarrow$Chinese show consistent gains over RL baselines, and for English$\to$Turkish, performance on COMETKiwi is comparable to advanced LLM-based systems (DeepSeek-V3.2). Our code and a set of representative pretrained models are publicly available at \url{https://github.com/NJUNLP/peg-rl} and \url{https://huggingface.co/collections/DGME/pegrl}."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shen-etal-2026-pegrl">
<titleInfo>
<title>PEGRL: Improving Machine Translation by Post-Editing Guided Reinforcement Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yunzhi</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xue</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junlan</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shujian</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Reinforcement learning (RL) has shown strong promise for LLM-based machine translation, with recent methods such as GRPO demonstrating notable gains; nevertheless, translation-oriented RL remains challenged by high-variance policy gradients induced by Monte Carlo baselines, as well as a large trajectory space that favors global exploration over fine-grained local optimization. We introduce PEGRL, a two-stage RL framework that uses post-editing as an auxiliary task to stabilize training and guide overall optimization. At each step, translation outputs are sampled to construct post-editing inputs, enabling lower-variance gradients from the post-editing task to propagate through the entire framework while jointly supporting both global exploration and fine-grained local optimization. A task-specific weighting scheme further emphasizes the post-editing gradient, producing a biased yet more sample-efficient estimator. Experiments on EnglishFinnish, EnglishTurkish, and EnglishłeftrightarrowChinese show consistent gains over RL baselines, and for EnglishTurkish, performance on COMETKiwi is comparable to advanced LLM-based systems (DeepSeek-V3.2). Our code and a set of representative pretrained models are publicly available at https://github.com/NJUNLP/peg-rl and https://huggingface.co/collections/DGME/pegrl.</abstract>
<identifier type="citekey">shen-etal-2026-pegrl</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.851/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>17225</start>
<end>17242</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PEGRL: Improving Machine Translation by Post-Editing Guided Reinforcement Learning
%A Shen, Yunzhi
%A Zhou, Hao
%A Huang, Xin
%A Han, Xue
%A Feng, Junlan
%A Huang, Shujian
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F shen-etal-2026-pegrl
%X Reinforcement learning (RL) has shown strong promise for LLM-based machine translation, with recent methods such as GRPO demonstrating notable gains; nevertheless, translation-oriented RL remains challenged by high-variance policy gradients induced by Monte Carlo baselines, as well as a large trajectory space that favors global exploration over fine-grained local optimization. We introduce PEGRL, a two-stage RL framework that uses post-editing as an auxiliary task to stabilize training and guide overall optimization. At each step, translation outputs are sampled to construct post-editing inputs, enabling lower-variance gradients from the post-editing task to propagate through the entire framework while jointly supporting both global exploration and fine-grained local optimization. A task-specific weighting scheme further emphasizes the post-editing gradient, producing a biased yet more sample-efficient estimator. Experiments on EnglishFinnish, EnglishTurkish, and EnglishłeftrightarrowChinese show consistent gains over RL baselines, and for EnglishTurkish, performance on COMETKiwi is comparable to advanced LLM-based systems (DeepSeek-V3.2). Our code and a set of representative pretrained models are publicly available at https://github.com/NJUNLP/peg-rl and https://huggingface.co/collections/DGME/pegrl.
%U https://aclanthology.org/2026.findings-acl.851/
%P 17225-17242
Markdown (Informal)
[PEGRL: Improving Machine Translation by Post-Editing Guided Reinforcement Learning](https://aclanthology.org/2026.findings-acl.851/) (Shen et al., Findings 2026)
ACL