GOOD: Training-Free Guided Diffusion Sampling for Out-of-Distribution Detection
This work addresses the challenge of improving OOD detection in machine learning, particularly for realistic scenarios, by providing a more flexible and effective method for generating OOD samples, though it appears incremental as it builds upon existing diffusion model approaches.
The paper tackles the problem of generating diverse and semantically stable out-of-distribution (OOD) samples for OOD detection by proposing GOOD, a training-free framework that guides diffusion sampling using in-distribution classifiers, resulting in enhanced OOD detection performance as demonstrated through quantitative and qualitative analyses.
Recent advancements have explored text-to-image diffusion models for synthesizing out-of-distribution (OOD) samples, substantially enhancing the performance of OOD detection. However, existing approaches typically rely on perturbing text-conditioned embeddings, resulting in semantic instability and insufficient shift diversity, which limit generalization to realistic OOD. To address these challenges, we propose GOOD, a novel and flexible framework that directly guides diffusion sampling trajectories towards OOD regions using off-the-shelf in-distribution (ID) classifiers. GOOD incorporates dual-level guidance: (1) Image-level guidance based on the gradient of log partition to reduce input likelihood, drives samples toward low-density regions in pixel space. (2) Feature-level guidance, derived from k-NN distance in the classifier's latent space, promotes sampling in feature-sparse regions. Hence, this dual-guidance design enables more controllable and diverse OOD sample generation. Additionally, we introduce a unified OOD score that adaptively combines image and feature discrepancies, enhancing detection robustness. We perform thorough quantitative and qualitative analyses to evaluate the effectiveness of GOOD, demonstrating that training with samples generated by GOOD can notably enhance OOD detection performance.