FairCLIP's claimed fairness and performance gains over CLIP do not reproduce on two datasets, and its official implementation diverges from the paper's own formulation.
Examining gender and racial bias in large vision -- language models using a novel dataset of parallel images
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On the Reproducibility of "FairCLIP: Harnessing Fairness in Vision-Language Learning''
FairCLIP's claimed fairness and performance gains over CLIP do not reproduce on two datasets, and its official implementation diverges from the paper's own formulation.