THaMES: An End-to-End Tool for Hallucination Mitigation and Evaluation in Large Language Models

TL;DR

THaMES is an integrated framework that automates the evaluation and mitigation of hallucinations in large language models, improving their factual accuracy across various tasks with adaptable strategies.

Contribution

It introduces an end-to-end, standardized pipeline for hallucination detection and mitigation, combining automated test set generation, benchmarking, and multiple mitigation techniques.

Findings

Commercial models like GPT-4o benefit more from RAG.

Open-weight models like Llama-3.1-8B-Instruct gain from ICL.

PEFT improves Llama-3.1-8B-Instruct's performance.

Abstract

Hallucination, the generation of factually incorrect content, is a growing challenge in Large Language Models (LLMs). Existing detection and mitigation methods are often isolated and insufficient for domain-specific needs, lacking a standardized pipeline. This paper introduces THaMES (Tool for Hallucination Mitigations and EvaluationS), an integrated framework and library addressing this gap. THaMES offers an end-to-end solution for evaluating and mitigating hallucinations in LLMs, featuring automated test set generation, multifaceted benchmarking, and adaptable mitigation strategies. It automates test set creation from any corpus, ensuring high data quality, diversity, and cost-efficiency through techniques like batch processing, weighted sampling, and counterfactual validation. THaMES assesses a model's ability to detect and reduce hallucinations across various tasks, including text generation and binary classification, applying optimal mitigation strategies like In-Context Learning (ICL), Retrieval Augmented Generation (RAG), and Parameter-Efficient Fine-tuning (PEFT). Evaluations of state-of-the-art LLMs using a knowledge base of academic papers, political news, and Wikipedia reveal that commercial models like GPT-4o benefit more from RAG than ICL, while open-weight models like Llama-3.1-8B-Instruct and Mistral-Nemo gain more from ICL. Additionally, PEFT significantly enhances the performance of Llama-3.1-8B-Instruct in both evaluation tasks.