Applying variational Bayesian inference to the CLAP linear probe adapter improves calibration and high-confidence coverage in CLIP few-shot classification, with a modest accuracy trade-off.
Turrisi da Costa, Cees G.M
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
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
BayesAdapter: enhanced uncertainty estimation in CLIP few-shot adaptation
Applying variational Bayesian inference to the CLAP linear probe adapter improves calibration and high-confidence coverage in CLIP few-shot classification, with a modest accuracy trade-off.