MIST adapts CLIP with two Gaussian-sampled prompts per class and improves few-shot accuracy under extreme domain shift on four benchmarks.
A broader study of cross-domain few- shot learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multiple Stochastic Prompt Tuning for Few-shot Adaptation under Extreme Domain Shift
MIST adapts CLIP with two Gaussian-sampled prompts per class and improves few-shot accuracy under extreme domain shift on four benchmarks.