Scaling Vision-Language Models Is Not Enough to Mitigate Bias
Ranking
Overall
75
Content
90
Popularity
40
Observed public metrics from 1 member.
Merged summary
TL;DR - A study of 194 vision-language models finds that increasing model scale alone does little to mitigate complex biases. Training-data quality is more consistently associated with better worst-group performance.
- Scale-performance correlation drops from ρ=0.68 on ImageNet to ρ=0.48 on CelebA and ρ=0.05 on UrbanCars.
- Curated training data improves worst-group accuracy by up to 25% over similarly scaled uncurated data.
- Architectural effects, including patch size and image resolution, vary by bias type, benchmark, and spatial distribution.
Sources (1)
Scaling Vision-Language Models Is Not Enough to Mitigate Bias
Public signals
Semantic Scholar citations 0 · Semantic Scholar influential citations 0
TL;DR - A study of 194 vision-language models finds that increasing model scale alone does little to mitigate complex biases. Training-data quality is more consistently associated with better worst-group performance.
- Scale-performance correlation drops from ρ=0.68 on ImageNet to ρ=0.48 on CelebA and ρ=0.05 on UrbanCars.
- Curated training data improves worst-group accuracy by up to 25% over similarly scaled uncurated data.
- Architectural effects, including patch size and image resolution, vary by bias type, benchmark, and spatial distribution.