OmiAD: One-Step Adaptive Masked Diffusion Model for Multi-class Anomaly Detection via Adversarial Distillation

Yaoxuan Feng, Wenchao Chen, Yuxin Li, Bo Chen, Yubiao Wang +3 more
2/14/2026

No official code found. Best implementation: caoyunkang/AdaCLIP (304 stars, MIT, last commit 2025-07-07). 1 more below.

Semantic Scholar

Code Implementations(2)

caoyunkang/AdaCLIPCommunity73%
MIT

[ECCV2024] The Official Implementation for ''AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detection''

30419Shell, PythonJul 8, 20241 years agoMIT
luolundashu/OmiADCommunity59%

OmiAD: One-Step Adaptive Masked Diffusion Model for Multi-class Anomaly Detection via Adversarial Distillation(ICML 2025)

180May 7, 20251 years ago

Cite this paper

@article{feng2026omiad,
  title  = {OmiAD: One-Step Adaptive Masked Diffusion Model for Multi-class Anomaly Detection via Adversarial Distillation},
  author = {Yaoxuan Feng and Wenchao Chen and Yuxin Li and Bo Chen and Yubiao Wang and Zixuan Zhao and Hongwei Liu and Mingyuan Zhou},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:a8494332714aa6b8df3b6ddb20f3141808cd8ca7},
  journal = {ICML 2025 2025}
}

Discussion