Research map / Label-efficient and robust learning / OOD and anomaly detection
KLIP: Localized Distribution Shift Detection via KL-Divergence with Diffusion Priors in Inverse Problems
Alireza Kheirandish, Jihoon Hong, Sara Fridovich-Keil
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add block restriction to raw KL (ablation) | D_KL, Predictor-Corrector · CT dataset-level star | AUC: 0.54 -> 0.85 | Improved |
| add block restriction to raw KL (ablation) | D_KL, Predictor-Corrector · CT dataset-level tumor | AUC: 0.60 -> 0.65 | Improved |
| add block restriction to raw KL (ablation) | D_KL, Predictor-Corrector · CT image-level tumor | AUC: 0.86 -> 0.91 | Improved |
| add timestep restriction to raw KL (ablation) | D_KL, Predictor-Corrector · CT dataset-level tumor | AUC: 0.60 -> 0.78 | Improved |
| add timestep restriction to raw KL (ablation) | D_KL, Predictor-Corrector · CT dataset-level star | AUC: 0.54 -> 0.57 | No clear effect |
| add timestep restriction on top of block restriction (ablation) | KLIP, Predictor-Corrector · CT image-level tumor | AUC: 0.91 -> 0.88 | Hurt |
| add timestep restriction on top of block restriction (ablation) | KLIP, Predictor-Corrector · CT dataset-level tumor | AUC: 0.65 -> 0.78 | Improved |
| tune hyperparameters on tumor set instead of star set (ablation) | KLIP, Predictor-Corrector · CT dataset-level star | AUC: 0.86 -> 0.68 | Hurt |
| tune hyperparameters on tumor set instead of star set (ablation) | KLIP, Predictor-Corrector · CT dataset-level tumor | AUC: 0.78 -> 0.87 | Improved |
| increase Monte Carlo samples from 1 to 8 (ablation) | KLIP · CelebA scar image-level | AUC: 0.813 -> 0.904 | Improved |
| increase Monte Carlo samples from 8 to 32 (ablation) | KLIP · CelebA scar image-level | AUC: 0.904 -> 0.909 | No clear effect |
| mismatched reconstruction blur kernel sigma 3 instead of matched 9 (ablation) | KLIP · CelebA scar image-level | AUC: 0.904 -> 0.809 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- SimpleNet — SimpleNet: A Simple Network for Image Anomaly Detection and Localization (CVPR 2023) · better on 2 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- GOOD: Training-Free Guided Diffusion Sampling for Out-of-Distribution Detection (NeurIPS 2025 · similarity 0.88)
- Rethinking the Evaluation of Out-of-Distribution Detection: A Sorites Paradox (NeurIPS 2024 · similarity 0.86)
- ImageNet-OOD: Deciphering Modern Out-of-Distribution Detection Algorithms (ICLR 2024 · similarity 0.86)
- Backbone-Equated Diffusion OOD via Sparse Internal Snapshots (NeurIPS 2026 · similarity 0.85)
- Diffusion Priors for Variational Likelihood Estimation and Image Denoising (NeurIPS 2024 · similarity 0.85)
- Diffusion-based Layer-wise Semantic Reconstruction for Unsupervised Out-of-Distribution Detection (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
