Beyond Shortcuts: Mitigating Visual Illusions in Frozen VLMs via Qualitative Reasoning
超越捷径:通过定性推理缓解冻结VLM中的视觉错觉
机构 * Institute of Artificial Intelligence, Hefei Comprehensive National Science Center(人工智能研究院,合肥国家科学中心) ; Anhui Polytechnic University(安徽理工大学) ; Hefei University of Technology(合肥工业大学) ; Anhui University(安徽大学) ; IGS, Imperial College London(帝国理工学院伦敦分校)
专题命中 复杂问题求解 :reasoning(title,abstract)
AI总结 本文提出SQI框架,通过定性约束提升冻结VLM的视觉 grounding,解决视觉错觉问题,实验显示在DataCV 2026挑战中表现优异,提升准确率并提供更好的可解释性。
Comments 4 pages, 2 figures, and 1 table. This is a methodology paper for the DataCV 2026 Challenge (CVPR Workshops), Task 1, where our method ranked 2nd