Unsupervised Thematic Context Discovery for Explainable AI in Fact Verification: Advancing the CARAG Framework

Manju Vallayil, Parma Nand, Wei Qi Yan, Héctor Allende-Cid, Héctor Allende-Cid, Héctor Allende-Cid

2025

Abstract

This paper introduces CARAG-u, an unsupervised extension of the Context-Aware Retrieval Augmented Generation (CARAG) framework, designed to advance explainability in Automated Fact Verification (AFV) architectures. Unlike its predecessor, CARAG-u eliminates reliance on predefined thematic annotations and claim-evidence pair labels, by dynamically deriving thematic clusters and evidence pools from unstructured datasets. This innovation enables CARAG-u to balance local and global perspectives in evidence retrieval and explanation generation. We benchmark CARAG-u against Retrieval Augmented Generation (RAG) and compare it with CARAG, highlighting its unsupervised adaptability while maintaining a competitive performance. Evaluations on the FactVer dataset demonstrate CARAG-u’s ability to generate thematically coherent and context-sensitive post-hoc explanations, advancing Explainable AI in AFV. The implementation of CARAG-u, including all dependencies, is publicly available to ensure reproducibility and support further research.

Paper Citation

in Harvard Style

Vallayil M., Nand P., Yan W. and Allende-Cid H. (2025). Unsupervised Thematic Context Discovery for Explainable AI in Fact Verification: Advancing the CARAG Framework. In Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR; ISBN , SciTePress, pages 67-77. DOI: 10.5220/0013683400004000

in Bibtex Style

@conference{kdir25,
author={Manju Vallayil and Parma Nand and Wei Yan and Héctor Allende-Cid},
title={Unsupervised Thematic Context Discovery for Explainable AI in Fact Verification: Advancing the CARAG Framework},
booktitle={Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR},
year={2025},
pages={67-77},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013683400004000},
isbn={},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR
TI - Unsupervised Thematic Context Discovery for Explainable AI in Fact Verification: Advancing the CARAG Framework
SN -
AU - Vallayil M.
AU - Nand P.
AU - Yan W.
AU - Allende-Cid H.
PY - 2025
SP - 67
EP - 77
DO - 10.5220/0013683400004000
PB - SciTePress