Vision-Language Models Performing Zero-Shot Tasks Exhibit Disparities Between Gender Groups | IEEE Conference Publication | IEEE Xplore