An unlabeled test-time adaptation method updates concept vectors, the linear predictor, and adds residual concepts so that concept bottleneck classifiers stay accurate under distribution shifts.
Network dissection: Quantifying interpretability of deep visual representations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adaptive Concept Bottleneck for Foundation Models Under Distribution Shifts
An unlabeled test-time adaptation method updates concept vectors, the linear predictor, and adds residual concepts so that concept bottleneck classifiers stay accurate under distribution shifts.