@inproceedings{wang-etal-2025-separate,
title = "Separate the Wheat from the Chaff: Winnowing Down Divergent Views in Retrieval Augmented Generation",
author = "Wang, Song and
Chen, Zihan and
Wang, Peng and
Wei, Zhepei and
Tan, Zhen and
Meng, Yu and
Shen, Cong and
Li, Jundong",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-main.587/",
pages = "11626--11642",
ISBN = "979-8-89176-332-6",
abstract = "Retrieval-augmented generation (RAG) addresses the limitation of large language models (LLMs) in achieving up-to-date information by integrating external knowledge sources, but it is hindered by noisy or irrelevant retrieved data, leading to reduced accuracy. Additionally, most RAG methods rely on task-specific supervision, reducing their adaptability across domains. To overcome these challenges, we propose WinnowRAG, a novel multi-agent debate-based RAG framework. WinnowRAG operates in two stages: in Stage I, query-aware clustering groups similar documents, with each cluster assigned to an LLM agent for generating personalized responses. A critic LLM then consolidates these answers, forming super-agents. In Stage II, the super-agents engage in a structured discussion to filter out incorrect or irrelevant information, ensuring only relevant knowledge is used for final response generation. Crucially, WinnowRAG is unsupervised and leverages pretrained LLMs without requiring fine-tuning, making it easily adaptable to various tasks. The experiments on various realistic datasets demonstrate the effectiveness of WinnowRAG over state-of-the-art baselines."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2025-separate">
<titleInfo>
<title>Separate the Wheat from the Chaff: Winnowing Down Divergent Views in Retrieval Augmented Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Song</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zihan</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peng</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhepei</namePart>
<namePart type="family">Wei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhen</namePart>
<namePart type="family">Tan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cong</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jundong</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-332-6</identifier>
</relatedItem>
<abstract>Retrieval-augmented generation (RAG) addresses the limitation of large language models (LLMs) in achieving up-to-date information by integrating external knowledge sources, but it is hindered by noisy or irrelevant retrieved data, leading to reduced accuracy. Additionally, most RAG methods rely on task-specific supervision, reducing their adaptability across domains. To overcome these challenges, we propose WinnowRAG, a novel multi-agent debate-based RAG framework. WinnowRAG operates in two stages: in Stage I, query-aware clustering groups similar documents, with each cluster assigned to an LLM agent for generating personalized responses. A critic LLM then consolidates these answers, forming super-agents. In Stage II, the super-agents engage in a structured discussion to filter out incorrect or irrelevant information, ensuring only relevant knowledge is used for final response generation. Crucially, WinnowRAG is unsupervised and leverages pretrained LLMs without requiring fine-tuning, making it easily adaptable to various tasks. The experiments on various realistic datasets demonstrate the effectiveness of WinnowRAG over state-of-the-art baselines.</abstract>
<identifier type="citekey">wang-etal-2025-separate</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-main.587/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>11626</start>
<end>11642</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Separate the Wheat from the Chaff: Winnowing Down Divergent Views in Retrieval Augmented Generation
%A Wang, Song
%A Chen, Zihan
%A Wang, Peng
%A Wei, Zhepei
%A Tan, Zhen
%A Meng, Yu
%A Shen, Cong
%A Li, Jundong
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F wang-etal-2025-separate
%X Retrieval-augmented generation (RAG) addresses the limitation of large language models (LLMs) in achieving up-to-date information by integrating external knowledge sources, but it is hindered by noisy or irrelevant retrieved data, leading to reduced accuracy. Additionally, most RAG methods rely on task-specific supervision, reducing their adaptability across domains. To overcome these challenges, we propose WinnowRAG, a novel multi-agent debate-based RAG framework. WinnowRAG operates in two stages: in Stage I, query-aware clustering groups similar documents, with each cluster assigned to an LLM agent for generating personalized responses. A critic LLM then consolidates these answers, forming super-agents. In Stage II, the super-agents engage in a structured discussion to filter out incorrect or irrelevant information, ensuring only relevant knowledge is used for final response generation. Crucially, WinnowRAG is unsupervised and leverages pretrained LLMs without requiring fine-tuning, making it easily adaptable to various tasks. The experiments on various realistic datasets demonstrate the effectiveness of WinnowRAG over state-of-the-art baselines.
%U https://aclanthology.org/2025.emnlp-main.587/
%P 11626-11642
Markdown (Informal)
[Separate the Wheat from the Chaff: Winnowing Down Divergent Views in Retrieval Augmented Generation](https://aclanthology.org/2025.emnlp-main.587/) (Wang et al., EMNLP 2025)
ACL
- Song Wang, Zihan Chen, Peng Wang, Zhepei Wei, Zhen Tan, Yu Meng, Cong Shen, and Jundong Li. 2025. Separate the Wheat from the Chaff: Winnowing Down Divergent Views in Retrieval Augmented Generation. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 11626–11642, Suzhou, China. Association for Computational Linguistics.