@inproceedings{chen-etal-2026-jury,
title = "{JURY}-{RL}: Votes Propose, Proofs Dispose for Label-Free {RLVR}",
author = "Chen, Xinjie and
Fu, Biao and
Jing, Wu and
Chen, Guoxin and
Liu, Xinggao and
Liu, Dayiheng and
Liao, Minpeng",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.2466/",
pages = "55987--56014",
ISBN = "979-8-89176-459-0",
abstract = "Reinforcement learning with verifiable rewards (RLVR) enhances reasoning in large language models (LLMs) but typically depends on human-annotated answers or curated reward specifications. Label-free alternatives such as majority voting or LLM-as-a-Judge remove this cost but introduce false positives that destabilize training. We introduce JURY-RL, a label-free RLVR framework that decouples answer proposal from reward disposal: rollouts matching the plurality-voted answer receive positive reward only when that answer is verified in Lean. When verification is inconclusive, ResZero (Residual-Zero) discards the unverified plurality and redistributes a zero-mean, variance-preserving signal over the residual answers, keeping the optimization gradient stable without reinforcing unverifiable consensus. Across three backbones trained on math data, JURY-RL outperforms label-free baselines on mathematical reasoning, transfers to code generation and general benchmarks, and surpasses a majority-vote baseline in the test-time RL (TTRL) setting; it matches supervised ground-truth training on pass@1 and beats it on pass@k and response diversity."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chen-etal-2026-jury">
<titleInfo>
<title>JURY-RL: Votes Propose, Proofs Dispose for Label-Free RLVR</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xinjie</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Biao</namePart>
<namePart type="family">Fu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wu</namePart>
<namePart type="family">Jing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guoxin</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinggao</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dayiheng</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minpeng</namePart>
<namePart type="family">Liao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Reinforcement learning with verifiable rewards (RLVR) enhances reasoning in large language models (LLMs) but typically depends on human-annotated answers or curated reward specifications. Label-free alternatives such as majority voting or LLM-as-a-Judge remove this cost but introduce false positives that destabilize training. We introduce JURY-RL, a label-free RLVR framework that decouples answer proposal from reward disposal: rollouts matching the plurality-voted answer receive positive reward only when that answer is verified in Lean. When verification is inconclusive, ResZero (Residual-Zero) discards the unverified plurality and redistributes a zero-mean, variance-preserving signal over the residual answers, keeping the optimization gradient stable without reinforcing unverifiable consensus. Across three backbones trained on math data, JURY-RL outperforms label-free baselines on mathematical reasoning, transfers to code generation and general benchmarks, and surpasses a majority-vote baseline in the test-time RL (TTRL) setting; it matches supervised ground-truth training on pass@1 and beats it on pass@k and response diversity.</abstract>
<identifier type="citekey">chen-etal-2026-jury</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2466/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>55987</start>
<end>56014</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T JURY-RL: Votes Propose, Proofs Dispose for Label-Free RLVR
%A Chen, Xinjie
%A Fu, Biao
%A Jing, Wu
%A Chen, Guoxin
%A Liu, Xinggao
%A Liu, Dayiheng
%A Liao, Minpeng
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F chen-etal-2026-jury
%X Reinforcement learning with verifiable rewards (RLVR) enhances reasoning in large language models (LLMs) but typically depends on human-annotated answers or curated reward specifications. Label-free alternatives such as majority voting or LLM-as-a-Judge remove this cost but introduce false positives that destabilize training. We introduce JURY-RL, a label-free RLVR framework that decouples answer proposal from reward disposal: rollouts matching the plurality-voted answer receive positive reward only when that answer is verified in Lean. When verification is inconclusive, ResZero (Residual-Zero) discards the unverified plurality and redistributes a zero-mean, variance-preserving signal over the residual answers, keeping the optimization gradient stable without reinforcing unverifiable consensus. Across three backbones trained on math data, JURY-RL outperforms label-free baselines on mathematical reasoning, transfers to code generation and general benchmarks, and surpasses a majority-vote baseline in the test-time RL (TTRL) setting; it matches supervised ground-truth training on pass@1 and beats it on pass@k and response diversity.
%U https://aclanthology.org/2026.emnlp-main.2466/
%P 55987-56014
Markdown (Informal)
[JURY-RL: Votes Propose, Proofs Dispose for Label-Free RLVR](https://aclanthology.org/2026.emnlp-main.2466/) (Chen et al., EMNLP 2026)
ACL
- Xinjie Chen, Biao Fu, Wu Jing, Guoxin Chen, Xinggao Liu, Dayiheng Liu, and Minpeng Liao. 2026. JURY-RL: Votes Propose, Proofs Dispose for Label-Free RLVR. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 55987–56014, Budapest, Hungary. Association for Computational Linguistics.