An Explainable KG-RAG-Based Approach to Evidence-Based Fake News Detection Using LLMs

Jonathan John Thomas; Radu-Casian Mihailescu

doi:10.1609/aaaiss.v6i1.36047

An Explainable KG-RAG-Based Approach to Evidence-Based Fake News Detection Using LLMs

Jonathan John Thomas
, Radu-Casian Mihailescu

School of Mathematical & Computer Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The advent of the Internet and social media has led to the rapid proliferation of fake news. Current state-of-the-art approaches for evidence-based fake news detection primarily utilize vector-based Retrieval Augmented Generation (RAG) systems. Recent studies have proposed RAG systems that outperform vector-based RAG systems by modeling the document store as a Knowledge Graph (KG). In this work, we investigated the performance of a KG-RAG-based approach for evidence-based fake news detection on the AVeriTeC dataset.

Original language	English
Title of host publication	Proceedings of the AAAI Symposium Series
Publisher	AAAI Press
Pages	152-154
Number of pages	3
ISBN (Electronic)	9781577358992
ISBN (Print)	1577358996
DOIs	https://doi.org/10.1609/aaaiss.v6i1.36047
Publication status	Published - 1 Aug 2025
Event	AAAI 2025 Summer Symposium: Context-Awareness in Cyber-Physical Systems - Heriot-Watt University Dubai, Dubai, United Arab Emirates Duration: 20 May 2025 → 22 May 2025 https://sites.google.com/view/cyber-physical-systems https://haic2025.com/

Publication series

Name	Proceedings of the 2025 AAAI Summer Symposium Series
Publisher	AAAI
Number	1
Volume	6
ISSN (Print)	2994-4317

Conference

Conference	AAAI 2025 Summer Symposium
Country/Territory	United Arab Emirates
City	Dubai
Period	20/05/25 → 22/05/25
Internet address	https://sites.google.com/view/cyber-physical-systems https://haic2025.com/

Access to Document

10.1609/aaaiss.v6i1.36047

https://ojs.aaai.org/index.php/AAAI-SS/article/view/36047

Cite this

@inproceedings{7e46b53a0ba941ec95166f49dd66a1c7,

title = "An Explainable KG-RAG-Based Approach to Evidence-Based Fake News Detection Using LLMs",

abstract = "The advent of the Internet and social media has led to the rapid proliferation of fake news. Current state-of-the-art approaches for evidence-based fake news detection primarily utilize vector-based Retrieval Augmented Generation (RAG) systems. Recent studies have proposed RAG systems that outperform vector-based RAG systems by modeling the document store as a Knowledge Graph (KG). In this work, we investigated the performance of a KG-RAG-based approach for evidence-based fake news detection on the AVeriTeC dataset.",

author = "Thomas, \{Jonathan John\} and Radu-Casian Mihailescu",

year = "2025",

month = aug,

day = "1",

doi = "10.1609/aaaiss.v6i1.36047",

language = "English",

isbn = "1577358996",

series = "Proceedings of the 2025 AAAI Summer Symposium Series",

publisher = "AAAI Press",

number = "1",

pages = "152--154",

booktitle = "Proceedings of the AAAI Symposium Series",

note = "AAAI 2025 Summer Symposium : Context-Awareness in Cyber-Physical Systems ; Conference date: 20-05-2025 Through 22-05-2025",

url = "https://sites.google.com/view/cyber-physical-systems, https://haic2025.com/",

}

TY - GEN

T1 - An Explainable KG-RAG-Based Approach to Evidence-Based Fake News Detection Using LLMs

AU - Thomas, Jonathan John

AU - Mihailescu, Radu-Casian

PY - 2025/8/1

Y1 - 2025/8/1

N2 - The advent of the Internet and social media has led to the rapid proliferation of fake news. Current state-of-the-art approaches for evidence-based fake news detection primarily utilize vector-based Retrieval Augmented Generation (RAG) systems. Recent studies have proposed RAG systems that outperform vector-based RAG systems by modeling the document store as a Knowledge Graph (KG). In this work, we investigated the performance of a KG-RAG-based approach for evidence-based fake news detection on the AVeriTeC dataset.

AB - The advent of the Internet and social media has led to the rapid proliferation of fake news. Current state-of-the-art approaches for evidence-based fake news detection primarily utilize vector-based Retrieval Augmented Generation (RAG) systems. Recent studies have proposed RAG systems that outperform vector-based RAG systems by modeling the document store as a Knowledge Graph (KG). In this work, we investigated the performance of a KG-RAG-based approach for evidence-based fake news detection on the AVeriTeC dataset.

U2 - 10.1609/aaaiss.v6i1.36047

DO - 10.1609/aaaiss.v6i1.36047

M3 - Conference contribution

SN - 1577358996

T3 - Proceedings of the 2025 AAAI Summer Symposium Series

SP - 152

EP - 154

BT - Proceedings of the AAAI Symposium Series

PB - AAAI Press

T2 - AAAI 2025 Summer Symposium

Y2 - 20 May 2025 through 22 May 2025

ER -

An Explainable KG-RAG-Based Approach to Evidence-Based Fake News Detection Using LLMs

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this