Hanxun Huang Sarah M. Erfani Yige Li Xingjun Ma James Bailey 0001 Detecting Backdoor Samples in Contrastive Language Image Pretraining. 2025 February abs/2502.01385 CoRR https://doi.org/10.48550/arXiv.2502.01385 db/journals/corr/corr2502.html#abs-2502-01385 streams/journals/corr