@inproceedings{fang-etal-2026-sparkle,
title = "{SPARKLE}: A Structured and Plug-and-play Agentic Retrieval Policy for Adaptive {RAG} Models",
author = "Fang, Jinyuan and
Meng, Zaiqiao and
Macdonald, Craig",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.1793/",
pages = "38699--38719",
ISBN = "979-8-89176-390-6",
abstract = "Adaptive retrieval-augmented generation (RAG) models offer an effective approach for integrating external knowledge. However, existing methods either rely on frozen large language models (LLMs) without explicit supervision or require costly LLM finetuning. Therefore, we propose SPARKLE, a structured and plug-and-play agentic retrieval policy where an additional proxy model is introduced to control the retrieval process. The proxy model leverages knowledge graph-based reasoning to make retrieval decisions in a structured manner, while operating independently of the retriever and the LLM. This plug-and-play design allows SPARKLE to generalise across different retrievers and LLMs. SPARKLE is optimised via reinforcement learning (RL), treating the retriever and the LLM as part of the environment. To enable more effective exploration during RL training, we further introduce a binary tree-structured rollout strategy. Experiments on three in-domain and four out-of-domain QA benchmarks show that SPARKLE outperforms state-of-the-art adaptive RAG baselines, achieving average improvements of 9.17{\%} and 2.85{\%}, respectively."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="fang-etal-2026-sparkle">
<titleInfo>
<title>SPARKLE: A Structured and Plug-and-play Agentic Retrieval Policy for Adaptive RAG Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jinyuan</namePart>
<namePart type="family">Fang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zaiqiao</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Craig</namePart>
<namePart type="family">Macdonald</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>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</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">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Adaptive retrieval-augmented generation (RAG) models offer an effective approach for integrating external knowledge. However, existing methods either rely on frozen large language models (LLMs) without explicit supervision or require costly LLM finetuning. Therefore, we propose SPARKLE, a structured and plug-and-play agentic retrieval policy where an additional proxy model is introduced to control the retrieval process. The proxy model leverages knowledge graph-based reasoning to make retrieval decisions in a structured manner, while operating independently of the retriever and the LLM. This plug-and-play design allows SPARKLE to generalise across different retrievers and LLMs. SPARKLE is optimised via reinforcement learning (RL), treating the retriever and the LLM as part of the environment. To enable more effective exploration during RL training, we further introduce a binary tree-structured rollout strategy. Experiments on three in-domain and four out-of-domain QA benchmarks show that SPARKLE outperforms state-of-the-art adaptive RAG baselines, achieving average improvements of 9.17% and 2.85%, respectively.</abstract>
<identifier type="citekey">fang-etal-2026-sparkle</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.1793/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>38699</start>
<end>38719</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SPARKLE: A Structured and Plug-and-play Agentic Retrieval Policy for Adaptive RAG Models
%A Fang, Jinyuan
%A Meng, Zaiqiao
%A Macdonald, Craig
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F fang-etal-2026-sparkle
%X Adaptive retrieval-augmented generation (RAG) models offer an effective approach for integrating external knowledge. However, existing methods either rely on frozen large language models (LLMs) without explicit supervision or require costly LLM finetuning. Therefore, we propose SPARKLE, a structured and plug-and-play agentic retrieval policy where an additional proxy model is introduced to control the retrieval process. The proxy model leverages knowledge graph-based reasoning to make retrieval decisions in a structured manner, while operating independently of the retriever and the LLM. This plug-and-play design allows SPARKLE to generalise across different retrievers and LLMs. SPARKLE is optimised via reinforcement learning (RL), treating the retriever and the LLM as part of the environment. To enable more effective exploration during RL training, we further introduce a binary tree-structured rollout strategy. Experiments on three in-domain and four out-of-domain QA benchmarks show that SPARKLE outperforms state-of-the-art adaptive RAG baselines, achieving average improvements of 9.17% and 2.85%, respectively.
%U https://aclanthology.org/2026.acl-long.1793/
%P 38699-38719
Markdown (Informal)
[SPARKLE: A Structured and Plug-and-play Agentic Retrieval Policy for Adaptive RAG Models](https://aclanthology.org/2026.acl-long.1793/) (Fang et al., ACL 2026)
ACL