Tackling the Inherent Difficulty of Noise Filtering in RAG

TL;DR

This paper addresses the challenge of noise filtering in Retrieval-Augmented Generation by highlighting the inherent difficulty of filtering irrelevant documents and proposing a novel fine-tuning method to improve model robustness against noisy retrievals.

Contribution

We introduce a new fine-tuning approach that enhances LLMs' ability to distinguish relevant from irrelevant information in retrieved documents, improving robustness.

Findings

Significant performance gains across multiple benchmarks.

Standard fine-tuning is often ineffective against noisy retrievals.

Our method improves the model's ability to ignore irrelevant content.

Abstract

Retrieval-Augmented Generation (RAG) has become a widely adopted approach to enhance Large Language Models (LLMs) by incorporating external knowledge and reducing hallucinations. However, noisy or irrelevant documents are often introduced during RAG, potentially degrading performance and even causing hallucinated outputs. While various methods have been proposed to filter out such noise, we argue that identifying irrelevant information from retrieved content is inherently difficult and limited number of transformer layers can hardly solve this. Consequently, retrievers fail to filter out irrelevant documents entirely. Therefore, LLMs must be robust against such noise, but we demonstrate that standard fine-tuning approaches are often ineffective in enabling the model to selectively utilize relevant information while ignoring irrelevant content due to the structural constraints of attention patterns. To address this, we propose a novel fine-tuning method designed to enhance the model's ability to distinguish between relevant and irrelevant information within retrieved documents. Extensive experiments across multiple benchmarks show that our approach significantly improves the robustness and performance of LLMs.