URAG: A Benchmark for Uncertainty Quantification in Retrieval-Augmented Large Language Models

TL;DR

URAG introduces a benchmark for evaluating the uncertainty and reliability of retrieval-augmented large language models across multiple domains, addressing a gap in current correctness-focused assessments.

Contribution

This paper presents URAG, a new benchmark that measures uncertainty in RAG systems using conformal prediction, across diverse fields and retrieval methods.

Findings

Simple modular RAG methods outperform complex reasoning pipelines in accuracy-uncertainty trade-offs.

Retrieval noise can break the typical accuracy-uncertainty relationship.

No RAG approach is reliably effective across all domains.

Abstract

Retrieval-Augmented Generation (RAG) has emerged as a widely adopted approach for enhancing LLMs in scenarios that demand extensive factual knowledge. However, current RAG evaluations concentrate primarily on correctness, which may not fully capture the impact of retrieval on LLM uncertainty and reliability. To bridge this gap, we introduce URAG, a comprehensive benchmark designed to assess the uncertainty of RAG systems across various fields like healthcare, programming, science, math, and general text. By reformulating open-ended generation tasks into multiple-choice question answering, URAG allows for principled uncertainty quantification via conformal prediction. We apply the evaluation pipeline to 8 standard RAG methods, measuring their performance through both accuracy and prediction-set sizes based on LAC and APS metrics. Our analysis shows that (1) accuracy gains often coincide with reduced uncertainty, but this relationship breaks under retrieval noise; (2) simple modular RAG methods tend to offer better accuracy-uncertainty trade-offs than more complex reasoning pipelines; and (3) no single RAG approach is universally reliable across domains. We further show that (4) retrieval depth, parametric knowledge dependence, and exposure to confidence cues can amplify confident errors and hallucinations. Ultimately, URAG establishes a systematic benchmark for analyzing and enhancing the trustworthiness of retrieval-augmented systems. Our code is available on GitHub.