ArgRAG: Explainable retrieval augmented generation using quantitative bipolar argumentation

Zhu, Yuqicheng, Potyka, Nico, Hernández, Daniel, He, Yuan, Ding, Zifeng, Xiong, Bo, Zhou, Dongzhuoran, Kharlamov, Evgeny and Staab, Steffen 2025. ArgRAG: Explainable retrieval augmented generation using quantitative bipolar argumentation. Presented at: 19th International Conference on Neurosymbolic Learning and Reasoning (NeSy), USA, 8 - 10 September 2025. Proceedings of Machine Learning Research. , vol.284 PMLR,

Preview

PDF - Accepted Post-Print Version Download (3MB) | Preview

Abstract

Retrieval-Augmented Generation (RAG) enhances large language models by incorporating external knowledge, yet suffers from critical limitations in high-stakes domains—namely, sensitivity to noisy or contradictory evidence and opaque, stochastic decision-making. We propose ArgRAG, an explainable, and contestable alternative that replaces black-box reasoning with structured inference using a Quantitative Bipolar Argumentation Framework (QBAF). ArgRAG constructs a QBAF from retrieved documents and performs deterministic reasoning under gradual semantics. This allows faithfully explanaining and contesting decisions. Evaluated on two fact verification benchmarks, PubHealth and RAGuard, ArgRAG achieves strong accuracy while significantly improving transparency.

Item Type:	Conference or Workshop Item - published (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Schools > Computer Science & Informatics
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Publisher:	PMLR
Date of First Compliant Deposit:	23 September 2025
Date of Acceptance:	20 April 2025
Last Modified:	30 Jan 2026 14:31
URI:	https://orca.cardiff.ac.uk/id/eprint/181190

Actions (repository staff only)

Edit Item