作者位次 #3/8ACL 2026Poster· Natural Language Processing
M irror QA : Benchmarking Multimodal LLM s on Mirror-Orientation Reasoning
Jingping Liu · Xingchen Peng · Yan Zhou · Ziyan Liu · Jie Zhai · Ronghao Chen · Huacan Wang · Xiaofeng Jia
摘要
Multimodal large language models (MLLMs) have achieved remarkable progress in recent years, yet their ability to perform left–right reasoning in mirror contexts—a fundamental element of spatial cognition—remains underexplored. To address this gap, we introduce MirrorQA, a manually constructed ben...