Retrieval-Augmented Generation (RAG) systems have shown promise in enhancing the performance of Large Language Models (LLMs). However, these systems face challenges in effectively integrating external knowledge with the LLM's internal knowledge, often leading to issues with misleading or unhelpful information. This work aims to provide a systematic study on knowledge checking in RAG systems. We conduct a comprehensive analysis of LLM representation behaviors and demonstrate the significance of using representations in knowledge checking. Motivated by the findings, we further develop representation-based classifiers for knowledge filtering. We show substantial improvements in RAG performance, even when dealing with noisy knowledge databases. Our study provides new insights into leveraging LLM representations for enhancing the reliability and effectiveness of RAG systems.
@article{arxiv.2411.14572,
title = {Towards Knowledge Checking in Retrieval-augmented Generation: A Representation Perspective},
author = {Shenglai Zeng and Jiankun Zhang and Bingheng Li and Yuping Lin and Tianqi Zheng and Dante Everaert and Hanqing Lu and Hui Liu and Hui Liu and Yue Xing and Monica Xiao Cheng and Jiliang Tang},
journal= {arXiv preprint arXiv:2411.14572},
year = {2024}
}