Task-Oriented Multi-Modal Mutual Learning for Vision-Language Models | IEEE Conference Publication | IEEE Xplore