AI RESEARCH
Fine-Grained Class-Conditional Distribution Balancing for Debiased Learning
arXiv CS.CV
•
ArXi:2505.06831v2 Announce Type: replace Achieving group-robust generalization in the presence of spurious correlations remains a significant challenge, particularly when bias annotations are unavailable. Recent studies on Class-Conditional Distribution Balancing (CCDB) reveal that spurious correlations often stem from mismatches between the class-conditional and marginal distributions of bias attributes. They achieve promising results by addressing this issue through simple distribution matching in a bias-agnostic manner.