Enhancing Noise Robustness of Retrieval-Augmented Language Models with Adaptive Adversarial Training

TL;DR

This paper introduces RAAT, a novel adaptive adversarial training method for retrieval-augmented language models, significantly improving their robustness against various real-world retrieval noises.

Contribution

The study categorizes real-world retrieval noises and proposes RAAT, a dynamic adversarial training approach that enhances LLM robustness and noise recognition capabilities.

Findings

RAAT improves F1 and EM scores under diverse noise conditions.

Extensive experiments validate RAAT's effectiveness over baseline models.

Code and data are publicly released for reproducibility.

Abstract

Large Language Models (LLMs) exhibit substantial capabilities yet encounter challenges, including hallucination, outdated knowledge, and untraceable reasoning processes. Retrieval-augmented generation (RAG) has emerged as a promising solution, integrating knowledge from external databases to mitigate these challenges. However, inappropriate retrieved passages can potentially hinder the LLMs' capacity to generate comprehensive and high-quality responses. Prior RAG studies on the robustness of retrieval noises often confine themselves to a limited set of noise types, deviating from real-world retrieval environments and limiting practical applicability. In this study, we initially investigate retrieval noises and categorize them into three distinct types, reflecting real-world environments. We analyze the impact of these various retrieval noises on the robustness of LLMs. Subsequently, we propose a novel RAG approach known as Retrieval-augmented Adaptive Adversarial Training (RAAT). RAAT leverages adaptive adversarial training to dynamically adjust the model's training process in response to retrieval noises. Concurrently, it employs multi-task learning to ensure the model's capacity to internally recognize noisy contexts. Extensive experiments demonstrate that the LLaMA-2 7B model trained using RAAT exhibits significant improvements in F1 and EM scores under diverse noise conditions. For reproducibility, we release our code and data at: https://github.com/calubkk/RAAT.