Will Multimodal Models Be Dazzled by Multi-Image Visual Puzzles?May 5, 2026·Zhi Zhu,YaoQi Fan,Zhe Chen,Yue Cao,Yangzhou LiuTong Lu· 0 min read Cite URLTypeConference paperPublicationProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)Last updated on May 5, 2026AuthorsTong LuNanjing University← VMonarch: Efficient Video Diffusion Transformers with Structured Attention May 5, 2026Arbitrary Generative Video Interpolation Apr 24, 2026 →
