RaVL: Discovering and Mitigating Spurious Correlations in Fine-Tuned Vision-Language Models

Open in new window