New benchmarks based on ImageNet and WILDS show that no CLIP-based OOD detection method is consistently best once semantic and covariate shifts are combined.
Delving into out-of-distribution detection with vision-language representations,
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
unclear 1representative citing papers
citing papers explorer
-
A Benchmark and Evaluation for Real-World Out-of-Distribution Detection Using Vision-Language Models
New benchmarks based on ImageNet and WILDS show that no CLIP-based OOD detection method is consistently best once semantic and covariate shifts are combined.