A training-free pipeline that uses captions of high-loss images to drive a latent diffusion model to synthesize bias-conflict samples, improving classifier debiasing.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Debiasing Classifiers by Amplifying Bias with Latent Diffusion and Large Language Models
A training-free pipeline that uses captions of high-loss images to drive a latent diffusion model to synthesize bias-conflict samples, improving classifier debiasing.