@inproceedings{kang-etal-2026-curlrank,
title = "{C}u{RLR}ank: Adaptive Curriculum Reinforcement Learning for Reasoning-Intensive Reranking",
author = "Kang, Xin and
Zhao, Yunxiao and
Zhang, Jiayang and
Li, Juncai and
Yan, Zhichao and
Han, XiaoQi and
Li, Ru and
Zhang, Delvin Ce",
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.1868/",
pages = "42563--42592",
ISBN = "979-8-89176-459-0",
abstract = "Document ranking aims to order candidate documents based on their relevance to a given query. However, identifying relevant documents in complex real-world scenarios requires deep reasoning beyond surface-level matching, driving a shift from semantic matching to reasoning-intensive reranking. Since the latter requires integrating information across documents for complex reasoning, it presents two key challenges: (1) varying reasoning difficulty, where assessing document relevance to the query requires multi-step reasoning with different complexity; and (2) strong contextual sensitivity, where the relevance of a document is heavily influenced by other documents within the context window. To address these challenges, we propose CuRLRank, which consists of (1) CuRL, an adaptive curriculum reinforcement learning approach that enables LLMs to develop complex reasoning capabilities across varying levels of difficulty, and (2) IFRS, an iterative filtering ranking strategy to achieve efficient and robust ranking. Extensive experiments on BRIGHT and R2MED benchmarks demonstrate that our CuRLRank significantly outperforms baseline methods while achieving lower ranking latency."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kang-etal-2026-curlrank">
<titleInfo>
<title>CuRLRank: Adaptive Curriculum Reinforcement Learning for Reasoning-Intensive Reranking</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Kang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yunxiao</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiayang</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juncai</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhichao</namePart>
<namePart type="family">Yan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">XiaoQi</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ru</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Delvin</namePart>
<namePart type="given">Ce</namePart>
<namePart type="family">Zhang</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>Document ranking aims to order candidate documents based on their relevance to a given query. However, identifying relevant documents in complex real-world scenarios requires deep reasoning beyond surface-level matching, driving a shift from semantic matching to reasoning-intensive reranking. Since the latter requires integrating information across documents for complex reasoning, it presents two key challenges: (1) varying reasoning difficulty, where assessing document relevance to the query requires multi-step reasoning with different complexity; and (2) strong contextual sensitivity, where the relevance of a document is heavily influenced by other documents within the context window. To address these challenges, we propose CuRLRank, which consists of (1) CuRL, an adaptive curriculum reinforcement learning approach that enables LLMs to develop complex reasoning capabilities across varying levels of difficulty, and (2) IFRS, an iterative filtering ranking strategy to achieve efficient and robust ranking. Extensive experiments on BRIGHT and R2MED benchmarks demonstrate that our CuRLRank significantly outperforms baseline methods while achieving lower ranking latency.</abstract>
<identifier type="citekey">kang-etal-2026-curlrank</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1868/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>42563</start>
<end>42592</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CuRLRank: Adaptive Curriculum Reinforcement Learning for Reasoning-Intensive Reranking
%A Kang, Xin
%A Zhao, Yunxiao
%A Zhang, Jiayang
%A Li, Juncai
%A Yan, Zhichao
%A Han, XiaoQi
%A Li, Ru
%A Zhang, Delvin Ce
%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 kang-etal-2026-curlrank
%X Document ranking aims to order candidate documents based on their relevance to a given query. However, identifying relevant documents in complex real-world scenarios requires deep reasoning beyond surface-level matching, driving a shift from semantic matching to reasoning-intensive reranking. Since the latter requires integrating information across documents for complex reasoning, it presents two key challenges: (1) varying reasoning difficulty, where assessing document relevance to the query requires multi-step reasoning with different complexity; and (2) strong contextual sensitivity, where the relevance of a document is heavily influenced by other documents within the context window. To address these challenges, we propose CuRLRank, which consists of (1) CuRL, an adaptive curriculum reinforcement learning approach that enables LLMs to develop complex reasoning capabilities across varying levels of difficulty, and (2) IFRS, an iterative filtering ranking strategy to achieve efficient and robust ranking. Extensive experiments on BRIGHT and R2MED benchmarks demonstrate that our CuRLRank significantly outperforms baseline methods while achieving lower ranking latency.
%U https://aclanthology.org/2026.emnlp-main.1868/
%P 42563-42592
Markdown (Informal)
[CuRLRank: Adaptive Curriculum Reinforcement Learning for Reasoning-Intensive Reranking](https://aclanthology.org/2026.emnlp-main.1868/) (Kang et al., EMNLP 2026)
ACL
- Xin Kang, Yunxiao Zhao, Jiayang Zhang, Juncai Li, Zhichao Yan, XiaoQi Han, Ru Li, and Delvin Ce Zhang. 2026. CuRLRank: Adaptive Curriculum Reinforcement Learning for Reasoning-Intensive Reranking. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 42563–42592, Budapest, Hungary. Association for Computational Linguistics.