A Comprehensive Survey of Hallucination in Large Language, Image, Video and Audio Foundation Models

TL;DR

This survey reviews recent progress in understanding, detecting, and reducing hallucinations in large foundation models across multiple modalities, highlighting challenges and future directions for reliable AI systems.

Contribution

It provides a comprehensive framework and taxonomy for hallucination in multimodal foundation models, synthesizing recent detection and mitigation strategies.

Findings

Hallucination remains a major challenge across modalities.

Recent detection methods improve reliability of foundation models.

Mitigation strategies are emerging but need further development.

Abstract

The rapid advancement of foundation models (FMs) across language, image, audio, and video domains has shown remarkable capabilities in diverse tasks. However, the proliferation of FMs brings forth a critical challenge: the potential to generate hallucinated outputs, particularly in high-stakes applications. The tendency of foundation models to produce hallucinated content arguably represents the biggest hindrance to their widespread adoption in real-world scenarios, especially in domains where reliability and accuracy are paramount. This survey paper presents a comprehensive overview of recent developments that aim to identify and mitigate the problem of hallucination in FMs, spanning text, image, video, and audio modalities. By synthesizing recent advancements in detecting and mitigating hallucination across various modalities, the paper aims to provide valuable insights for researchers, developers, and practitioners. Essentially, it establishes a clear framework encompassing definition, taxonomy, and detection strategies for addressing hallucination in multimodal foundation models, laying the foundation for future research in this pivotal area.