BayesTTA adapts CLIP to gradually evolving distribution shifts by incrementally estimating class-conditional Gaussian statistics, selecting covariance structure via hypothesis testing, and refining predictions with Gaussian discriminant analysis plus zero-shot fusion.
Continual test-time domain adaptation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
BayesTTA: Continual-Temporal Test-Time Adaptation for Vision-Language Models via Gaussian Discriminant Analysis
BayesTTA adapts CLIP to gradually evolving distribution shifts by incrementally estimating class-conditional Gaussian statistics, selecting covariance structure via hypothesis testing, and refining predictions with Gaussian discriminant analysis plus zero-shot fusion.