A new conditioning network lets diffusion models generate fair medical images for demographic intersections never seen in training, beating reweighting baselines on chest X-ray and fundus data.
Adv.11, eads4593 (2025)
1 Pith paper cite this work, alongside 3 external citations. Polarity classification is still indexing.
1
Pith paper citing it
3
external citations · OpenAlex
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CompDiff: Hierarchical Compositional Diffusion for Fair and Zero-Shot Intersectional Medical Image Generation
A new conditioning network lets diffusion models generate fair medical images for demographic intersections never seen in training, beating reweighting baselines on chest X-ray and fundus data.