The Hallucinations Leaderboard -- An Open Effort to Measure Hallucinations in Large Language Models

TL;DR

This paper presents the Hallucinations Leaderboard, an open benchmark to quantitatively evaluate and compare the tendency of large language models to generate hallucinated, non-factual outputs across various NLP tasks.

Contribution

It introduces a comprehensive, open leaderboard with benchmarks for measuring hallucination tendencies in large language models, aiding in model evaluation and selection.

Findings

Different models vary significantly in hallucination rates.

The leaderboard provides a standardized way to assess factuality.

Insights help improve model reliability and trustworthiness.

Abstract

Large Language Models (LLMs) have transformed the Natural Language Processing (NLP) landscape with their remarkable ability to understand and generate human-like text. However, these models are prone to ``hallucinations'' -- outputs that do not align with factual reality or the input context. This paper introduces the Hallucinations Leaderboard, an open initiative to quantitatively measure and compare the tendency of each model to produce hallucinations. The leaderboard uses a comprehensive set of benchmarks focusing on different aspects of hallucinations, such as factuality and faithfulness, across various tasks, including question-answering, summarisation, and reading comprehension. Our analysis provides insights into the performance of different models, guiding researchers and practitioners in choosing the most reliable models for their applications.