Exploiting Diffusion Prior for Out-of-Distribution Detection

TL;DR

This paper introduces a new out-of-distribution detection method that combines diffusion models and CLIP features, enabling robust and scalable detection without needing class-specific labeled data, validated by extensive experiments.

Contribution

The paper proposes a novel OOD detection approach using diffusion models conditioned on CLIP features, eliminating the need for class-specific labeled data and improving detection accuracy.

Findings

Significantly improved OOD detection accuracy on benchmark datasets.

Method does not require class-specific labeled in-distribution data.

Demonstrates robustness and scalability in various scenarios.

Abstract

Out-of-distribution (OOD) detection is crucial for deploying robust machine learning models, especially in areas where security is critical. However, traditional OOD detection methods often fail to capture complex data distributions from large scale date. In this paper, we present a novel approach for OOD detection that leverages the generative ability of diffusion models and the powerful feature extraction capabilities of CLIP. By using these features as conditional inputs to a diffusion model, we can reconstruct the images after encoding them with CLIP. The difference between the original and reconstructed images is used as a signal for OOD identification. The practicality and scalability of our method is increased by the fact that it does not require class-specific labeled ID data, as is the case with many other methods. Extensive experiments on several benchmark datasets demonstrates the robustness and effectiveness of our method, which have significantly improved the detection accuracy.