RE-RAG: Improving Open-Domain QA Performance and Interpretability with Relevance Estimator in Retrieval-Augmented Generation

TL;DR

This paper introduces RE-RAG, a framework that enhances open-domain question answering by using a relevance estimator to improve retrieval relevance and interpretability, leading to better performance and user trust.

Contribution

The paper proposes a relevance estimator with confidence scores for retrieval-augmented generation, trained with weak supervision, improving QA performance and interpretability.

Findings

RE-RAG improves QA accuracy with relevance confidence scores.

Training RE with small generator enhances large language models.

Decoding strategies using confidence scores improve answer reliability.

Abstract

The Retrieval Augmented Generation (RAG) framework utilizes a combination of parametric knowledge and external knowledge to demonstrate state-of-the-art performance on open-domain question answering tasks. However, the RAG framework suffers from performance degradation when the query is accompanied by irrelevant contexts. In this work, we propose the RE-RAG framework, which introduces a relevance estimator (RE) that not only provides relative relevance between contexts as previous rerankers did, but also provides confidence, which can be used to classify whether given context is useful for answering the given question. We propose a weakly supervised method for training the RE simply utilizing question-answer data without any labels for correct contexts. We show that RE trained with a small generator (sLM) can not only improve the sLM fine-tuned together with RE but also improve previously unreferenced large language models (LLMs). Furthermore, we investigate new decoding strategies that utilize the proposed confidence measured by RE such as choosing to let the user know that it is "unanswerable" to answer the question given the retrieved contexts or choosing to rely on LLM's parametric knowledge rather than unrelated contexts.