@inproceedings{jiang-etal-2026-breaking,
title = "Breaking Consensus Bias: Unsupervised Reinforcement Learning for Machine Translation",
author = "Jiang, Shuting and
Song, Ran and
Zhang, Siqi and
Huang, Yuxin and
Gao, Shengxiang and
Yu, Zhengtao",
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.1042/",
pages = "20798--20812",
ISBN = "979-8-89176-395-1",
abstract = "Reinforcement learning (RL) excels in reasoning tasks with verifiable rewards, while its adaptation to machine translation (MT) remains challenging due to the lack of unique reward signals under multiple valid translations. Existing RL approaches for MT face either fixed references in supervised settings or the production of homogeneous references leading to mode collapse in unsupervised settings. Both limitations arise from ignoring entropy dynamics in RL-based MT. The core challenge is leveraging entropy for supervision construction and self-evolution. In this paper, we propose an Entropy-Driven Unsupervised RL for MT. Our framework integrates entropy-guided sampling for exploration, confidence-weighted label generation to transcend majority-voting bias, and uncertainty-aware optimization to prioritize high-entropy tokens. These mechanisms allow reward signals to co-evolve with model proficiency beyond fixed references. Experiments across multiple language pairs show our method outperforms supervised and unsupervised baselines by +0.63 and +2.52 average points, respectively. Our code is available at https://github.com/fortunatekiss/URLMT."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jiang-etal-2026-breaking">
<titleInfo>
<title>Breaking Consensus Bias: Unsupervised Reinforcement Learning for Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shuting</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ran</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siqi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuxin</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shengxiang</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhengtao</namePart>
<namePart type="family">Yu</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) excels in reasoning tasks with verifiable rewards, while its adaptation to machine translation (MT) remains challenging due to the lack of unique reward signals under multiple valid translations. Existing RL approaches for MT face either fixed references in supervised settings or the production of homogeneous references leading to mode collapse in unsupervised settings. Both limitations arise from ignoring entropy dynamics in RL-based MT. The core challenge is leveraging entropy for supervision construction and self-evolution. In this paper, we propose an Entropy-Driven Unsupervised RL for MT. Our framework integrates entropy-guided sampling for exploration, confidence-weighted label generation to transcend majority-voting bias, and uncertainty-aware optimization to prioritize high-entropy tokens. These mechanisms allow reward signals to co-evolve with model proficiency beyond fixed references. Experiments across multiple language pairs show our method outperforms supervised and unsupervised baselines by +0.63 and +2.52 average points, respectively. Our code is available at https://github.com/fortunatekiss/URLMT.</abstract>
<identifier type="citekey">jiang-etal-2026-breaking</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.1042/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>20798</start>
<end>20812</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Breaking Consensus Bias: Unsupervised Reinforcement Learning for Machine Translation
%A Jiang, Shuting
%A Song, Ran
%A Zhang, Siqi
%A Huang, Yuxin
%A Gao, Shengxiang
%A Yu, Zhengtao
%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 jiang-etal-2026-breaking
%X Reinforcement learning (RL) excels in reasoning tasks with verifiable rewards, while its adaptation to machine translation (MT) remains challenging due to the lack of unique reward signals under multiple valid translations. Existing RL approaches for MT face either fixed references in supervised settings or the production of homogeneous references leading to mode collapse in unsupervised settings. Both limitations arise from ignoring entropy dynamics in RL-based MT. The core challenge is leveraging entropy for supervision construction and self-evolution. In this paper, we propose an Entropy-Driven Unsupervised RL for MT. Our framework integrates entropy-guided sampling for exploration, confidence-weighted label generation to transcend majority-voting bias, and uncertainty-aware optimization to prioritize high-entropy tokens. These mechanisms allow reward signals to co-evolve with model proficiency beyond fixed references. Experiments across multiple language pairs show our method outperforms supervised and unsupervised baselines by +0.63 and +2.52 average points, respectively. Our code is available at https://github.com/fortunatekiss/URLMT.
%U https://aclanthology.org/2026.findings-acl.1042/
%P 20798-20812
Markdown (Informal)
[Breaking Consensus Bias: Unsupervised Reinforcement Learning for Machine Translation](https://aclanthology.org/2026.findings-acl.1042/) (Jiang et al., Findings 2026)
ACL