Large Language Models for Anomaly and Out-of-Distribution Detection: A Survey

2024-09-03

arXiv.org

AI Summary

Large language models (LLMs) exhibit weak generalization and poor interpretability in anomaly detection and out-of-distribution (OOD) sample identification. Method: This paper proposes the first two-dimensional classification framework centered on LLM functional roles—namely, *assistant* (generative aid) and *discriminator* (decision-oriented classifier)—departing from conventional statistical or supervised paradigms. It systematically integrates over one hundred LLM-adapted detection methods via synergistic techniques: prompt engineering, self-supervised representation analysis, uncertainty modeling, and multimodal alignment. Contribution/Results: The work distills four fundamental challenges and five scalable research directions, and establishes a dynamically updated, authoritative literature repository. Empirical outcomes demonstrate substantial improvements in methodological systematicity, interpretability, and cross-task transferability—providing both theoretical foundations and practical guidelines for trustworthy LLM deployment.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Detecting anomalies or out-of-distribution (OOD) samples is critical for maintaining the reliability and trustworthiness of machine learning systems. Recently, Large Language Models (LLMs) have demonstrated their effectiveness not only in natural language processing but also in broader applications due to their advanced comprehension and generative capabilities. The integration of LLMs into anomaly and OOD detection marks a significant shift from the traditional paradigm in the field. This survey focuses on the problem of anomaly and OOD detection under the context of LLMs. We propose a new taxonomy to categorize existing approaches into two classes based on the role played by LLMs. Following our proposed taxonomy, we further discuss the related work under each of the categories and finally discuss potential challenges and directions for future research in this field. We also provide an up-to-date reading list of relevant papers.

Problem

• Anomaly detection with Large Language Models
• Out-of-distribution sample identification using LLMs
• Surveying LLM integration in traditional detection paradigms

Innovation

• LLMs for anomaly detection
• New taxonomy classification
• Future research challenges