ALSA learns anchors in logit space and uses their influence on unlabeled samples to estimate model accuracy under distribution shift.
Are labels always necessary for classifier accuracy evaluation? In CVPR, 2021
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.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
ALSA: Anchors in Logit Space for Out-of-Distribution Accuracy Estimation
ALSA learns anchors in logit space and uses their influence on unlabeled samples to estimate model accuracy under distribution shift.