Evaluating RAG-Fusion with RAGElo: an Automated Elo-based Framework

TL;DR

This paper introduces RAGElo, an automated Elo-based framework leveraging LLMs for evaluating and ranking Retrieval-Augmented Generation systems in domain-specific QA tasks, addressing hallucination and benchmarking challenges.

Contribution

It presents a comprehensive, LLM-based evaluation framework for RAG systems, including synthetic data generation, LLM-as-judge, and Elo ranking, specifically applied to domain-specific QA.

Findings

RAGF outperforms RAG in Elo score and answer completeness.

RAGElo's rankings correlate positively with human judgments.

RAGF produces more complete and relevant answers.

Abstract

Challenges in the automated evaluation of Retrieval-Augmented Generation (RAG) Question-Answering (QA) systems include hallucination problems in domain-specific knowledge and the lack of gold standard benchmarks for company internal tasks. This results in difficulties in evaluating RAG variations, like RAG-Fusion (RAGF), in the context of a product QA task at Infineon Technologies. To solve these problems, we propose a comprehensive evaluation framework, which leverages Large Language Models (LLMs) to generate large datasets of synthetic queries based on real user queries and in-domain documents, uses LLM-as-a-judge to rate retrieved documents and answers, evaluates the quality of answers, and ranks different variants of Retrieval-Augmented Generation (RAG) agents with RAGElo's automated Elo-based competition. LLM-as-a-judge rating of a random sample of synthetic queries shows a moderate, positive correlation with domain expert scoring in relevance, accuracy, completeness, and precision. While RAGF outperformed RAG in Elo score, a significance analysis against expert annotations also shows that RAGF significantly outperforms RAG in completeness, but underperforms in precision. In addition, Infineon's RAGF assistant demonstrated slightly higher performance in document relevance based on MRR@5 scores. We find that RAGElo positively aligns with the preferences of human annotators, though due caution is still required. Finally, RAGF's approach leads to more complete answers based on expert annotations and better answers overall based on RAGElo's evaluation criteria.