作者位次 #4/8ACL 2026Poster· Natural Language Processing
How Do LLM s and VLM s Understand Viewpoint Rotation Without Vision? An Interpretability Study
Zhen Yang · Ping Jian · Zhongbin Guo · Zuming Zhang · Chengzhi Li · Yonghong Deng · Xinyue Zhang · Wenpeng Lu (Key Laboratory of Computing Power Network and Information Security, Ministry of Education, Shandong Computer Science Center (National Supercomputer Center in Jinan), Qilu University of Technology (Shandong Academy of Sciences)
摘要
Over the past year, spatial intelligence has drawn increasing attention. Many prior works study it from the perspective of visual-spatial intelligence, where models have access to visuospatial information from visual inputs. However, in the absence of visual information, whether linguistic intell...