Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback

TL;DR

This paper introduces LLM-Augmenter, a system that enhances large language models by grounding responses in external knowledge and using automated feedback to reduce hallucinations while maintaining response quality.

Contribution

It presents a modular system that improves LLM factuality and response accuracy through external knowledge integration and iterative prompt refinement.

Findings

Significantly reduces hallucinations in ChatGPT responses

Maintains fluency and informativeness of generated responses

Validated on dialog and question answering tasks

Abstract

Large language models (LLMs), such as ChatGPT, are able to generate human-like, fluent responses for many downstream tasks, e.g., task-oriented dialog and question answering. However, applying LLMs to real-world, mission-critical applications remains challenging mainly due to their tendency to generate hallucinations and their inability to use external knowledge. This paper proposes a LLM-Augmenter system, which augments a black-box LLM with a set of plug-and-play modules. Our system makes the LLM generate responses grounded in external knowledge, e.g., stored in task-specific databases. It also iteratively revises LLM prompts to improve model responses using feedback generated by utility functions, e.g., the factuality score of a LLM-generated response. The effectiveness of LLM-Augmenter is empirically validated on two types of scenarios, task-oriented dialog and open-domain question answering. LLM-Augmenter significantly reduces ChatGPT's hallucinations without sacrificing the fluency and informativeness of its responses. We make the source code and models publicly available.