Where do Large Vision-Language Models Look at when Answering Questions? Paper • 2503.13891 • Published 23 days ago • 8 • 2