diffusion-tt
Diffusion-TT A: Test-time Adaptation of Discriminative Models via Generative Feedback
Generative models have previously been used for test time adaptation of image classifiers and segmentors, by co-training the model under a joint discriminative task loss and a self-supervised image reconstruction loss, e.g., TTT -MAE [
Country:
- North America > United States > Pennsylvania > Allegheny County > Pittsburgh (0.04)
- Europe > United Kingdom (0.04)
Technology:
Exploring Structured Semantic Priors Underlying Diffusion Score for Test-time Adaptation
While Diffusion-TT A is competitive, there's still room to unleash the full potential of diffusion models. To achieve this, two key aspects warrant further exploration.
Country:
- North America > United States (0.09)
- Asia > China > Beijing > Beijing (0.04)
- Europe > Netherlands > South Holland > Delft (0.04)
Technology:
Exploring Structured Semantic Priors Underlying Diffusion Score for Test-time Adaptation
While Diffusion-TT A is competitive, there's still room to unleash the full potential of diffusion models. To achieve this, two key aspects warrant further exploration.
Country:
- North America > United States (0.09)
- Asia > China > Beijing > Beijing (0.04)
- Europe > Netherlands > South Holland > Delft (0.04)
Technology:
Country:
- North America > United States > Pennsylvania > Allegheny County > Pittsburgh (0.04)
- Europe > United Kingdom (0.04)
Technology: