DiMo-GUI: Advancing Test-time Scaling in GUI Grounding via Modality-Aware Visual Reasoning
机构 * University of California, Merced(加州大学梅尔德分校) ; The University of Queensland(昆士兰大学) ; vivo Mobile Communication Co., Ltd(vivo移动通信有限公司)
专题命中 视觉空间推理 :reasoning(title,abstract);分类 cs.AI
Comments EMNLP 2025 Main Conference