@inproceedings{qiang-etal-2025-exploring-knowledge,
title = "Exploring Knowledge Filtering for Retrieval-Augmented Discriminative Tasks",
author = "Qiang, Minjie and
Wang, Zhongqing and
Bao, Xiaoyi and
Ma, HaoYuan and
Li, Shoushan and
Zhou, Guodong",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.86/",
doi = "10.18653/v1/2025.findings-acl.86",
pages = "1716--1729",
ISBN = "979-8-89176-256-5",
abstract = "Retrieval-augmented methods have achieved remarkable advancements in alleviating the hallucination of large language models.Nevertheless, the introduction of external knowledge does not always lead to the expected improvement in model performance, as irrelevant or harmful information present in the retrieved knowledge can compromise the prediction process.To address these challenges, we propose a novel framework aimed at improving model performance by incorporating knowledge filtering and prediction fusion mechanisms.In particular, our approach first employs a perplexity-based annotation method to collect training data.Then, we design four distinct strategies to filter out harmful retrieved knowledge.Finally, we integrate the filtered knowledge to generate the final result via batch-wise predictions.We conduct extensive experiments across multiple discriminative task datasets to evaluate the proposed framework.The results demonstrate that our framework can significantly enhance the performance of models on discriminative tasks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="qiang-etal-2025-exploring-knowledge">
<titleInfo>
<title>Exploring Knowledge Filtering for Retrieval-Augmented Discriminative Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Minjie</namePart>
<namePart type="family">Qiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhongqing</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoyi</namePart>
<namePart type="family">Bao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">HaoYuan</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shoushan</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guodong</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-256-5</identifier>
</relatedItem>
<abstract>Retrieval-augmented methods have achieved remarkable advancements in alleviating the hallucination of large language models.Nevertheless, the introduction of external knowledge does not always lead to the expected improvement in model performance, as irrelevant or harmful information present in the retrieved knowledge can compromise the prediction process.To address these challenges, we propose a novel framework aimed at improving model performance by incorporating knowledge filtering and prediction fusion mechanisms.In particular, our approach first employs a perplexity-based annotation method to collect training data.Then, we design four distinct strategies to filter out harmful retrieved knowledge.Finally, we integrate the filtered knowledge to generate the final result via batch-wise predictions.We conduct extensive experiments across multiple discriminative task datasets to evaluate the proposed framework.The results demonstrate that our framework can significantly enhance the performance of models on discriminative tasks.</abstract>
<identifier type="citekey">qiang-etal-2025-exploring-knowledge</identifier>
<identifier type="doi">10.18653/v1/2025.findings-acl.86</identifier>
<location>
<url>https://aclanthology.org/2025.findings-acl.86/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>1716</start>
<end>1729</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Exploring Knowledge Filtering for Retrieval-Augmented Discriminative Tasks
%A Qiang, Minjie
%A Wang, Zhongqing
%A Bao, Xiaoyi
%A Ma, HaoYuan
%A Li, Shoushan
%A Zhou, Guodong
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Findings of the Association for Computational Linguistics: ACL 2025
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-256-5
%F qiang-etal-2025-exploring-knowledge
%X Retrieval-augmented methods have achieved remarkable advancements in alleviating the hallucination of large language models.Nevertheless, the introduction of external knowledge does not always lead to the expected improvement in model performance, as irrelevant or harmful information present in the retrieved knowledge can compromise the prediction process.To address these challenges, we propose a novel framework aimed at improving model performance by incorporating knowledge filtering and prediction fusion mechanisms.In particular, our approach first employs a perplexity-based annotation method to collect training data.Then, we design four distinct strategies to filter out harmful retrieved knowledge.Finally, we integrate the filtered knowledge to generate the final result via batch-wise predictions.We conduct extensive experiments across multiple discriminative task datasets to evaluate the proposed framework.The results demonstrate that our framework can significantly enhance the performance of models on discriminative tasks.
%R 10.18653/v1/2025.findings-acl.86
%U https://aclanthology.org/2025.findings-acl.86/
%U https://doi.org/10.18653/v1/2025.findings-acl.86
%P 1716-1729
Markdown (Informal)
[Exploring Knowledge Filtering for Retrieval-Augmented Discriminative Tasks](https://aclanthology.org/2025.findings-acl.86/) (Qiang et al., Findings 2025)
ACL