The GOOD framework adds a pre-trained General Knowledge Model to few-shot out-of-distribution detection, claims a provable general-specific balance that lowers generalization error, and reports superior benchmark results.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Generalized Few-Shot Out-of-Distribution Detection
The GOOD framework adds a pre-trained General Knowledge Model to few-shot out-of-distribution detection, claims a provable general-specific balance that lowers generalization error, and reports superior benchmark results.