Combining pre-trained Vision Transformers and CIDER for Out Of Domain Detection
This work addresses out-of-domain detection for industrial applications, but it is incremental as it combines existing methods without introducing new paradigms.
The paper tackled out-of-domain detection by evaluating pre-trained models, finding that Vision Transformers outperform CNNs and can be enhanced with CIDER refinement, achieving improved detection performance as a stronger baseline.
Out-of-domain (OOD) detection is a crucial component in industrial applications as it helps identify when a model encounters inputs that are outside the training distribution. Most industrial pipelines rely on pre-trained models for downstream tasks such as CNN or Vision Transformers. This paper investigates the performance of those models on the task of out-of-domain detection. Our experiments demonstrate that pre-trained transformers models achieve higher detection performance out of the box. Furthermore, we show that pre-trained ViT and CNNs can be combined with refinement methods such as CIDER to improve their OOD detection performance even more. Our results suggest that transformers are a promising approach for OOD detection and set a stronger baseline for this task in many contexts