Retrieval-Augmented Transformer Architecture for Cross-Domain Fake News Detection

Jyothilakshmi G Kava; Rajeshwari N

doi:10.36948/ijfmr.2026.v08i03.79319

Retrieval-Augmented Transformer Architecture for Cross-Domain Fake News Detection

Author(s)	Smt. Jyothilakshmi G Kava, Ms. Rajeshwari N
Country	India
Abstract	The rapid proliferation of fake news across digital platforms poses serious challenges to public trust, democratic processes, and information integrity. Although existing machine learning and deep learning models demonstrate high accuracy on domain-specific datasets, they often fail to generalize across unseen domains and are increasingly vulnerable to AI-generated misinformation. This paper proposes an AI-driven Hybrid Transformer–Retrieval Architecture for robust cross-domain fake news detection. The methodology integrates BERT-based contextual semantic encoding, retrieval-augmented factual verification using Dense Passage Retrieval (DPR) and Retrieval-Augmented Generation (RAG), credibility-based source scoring, stance detection, and a fusion-based decision layer. The model is trained on the Kaggle Fake News dataset and evaluated cross-domain on FakeNewsNet and GossipCop datasets. Experimental results show that the proposed model achieves 97.8% accuracy on Kaggle while maintaining 96.1% and 94.2% accuracy on FakeNewsNet and GossipCop respectively, representing a 27.4% improvement in cross-domain generalization over traditional machine learning and transformer-only approaches. The novelty of this work lies in unifying semantic understanding, factual grounding, and credibility reasoning into a single scalable framework for real-world misinformation detection.
Keywords	Fake news detection, cross-domain generalization, misinformation, BERT, transformers
Field	Computer Applications
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-05-26
DOI	https://doi.org/10.36948/ijfmr.2026.v08i03.79319

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 3 Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 2

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Retrieval-Augmented Transformer Architecture for Cross-Domain Fake News Detection

Share this