A radiology report can already answer a clinical question, so it is hard to tell whether a vision-language model also uses the image. ModaLens, a paired imag…
机构:MIT
来源:arXiv 2609.15635 | AI4Papers 论文推荐平台