Flow matching models have recently emerged as an efficient alternative to diffusion, especially for text-guided image generation and editing, offering faster inference through continuous-time dynamics. However, existing flow-based editors predominantly support global or single-instruction edits and struggle with multi-instance scenarios, where multiple parts of a reference input must be edited independently without semantic interference. We identify this limitation as a consequence of globally conditioned velocity fields and joint attention mechanisms, which entangle concurrent edits. To address this issue, we introduce Instance-Disentangled Attention, a mechanism that partitions joint attention operations, enforcing binding between instance-specific textual instructions and spatial regions during velocity field estimation. We evaluate our approach on both natural image editing and a newly introduced benchmark of text-dense infographics with region-level editing instructions. Experimental results demonstrate that our approach promotes edit disentanglement and locality while preserving global output coherence, enabling single-pass, instance-level editing.
StoryState: Agent-Based State Control for Consistent and Editable Storybooks
Birbhum Institute of Engineering and Technology(比尔布尔工程科技学院);Universiti Malaya(马来亚大学);City University of Hong Kong(香港城市大学);Wake Forest University(威克森林大学)
UltraViCo: Breaking Extrapolation Limits in Video Diffusion Transformers
Dept. of Comp. Sci. & Tech., BNRist Center, THU-Bosch ML Center, Tsinghua University(清华大学计算机科学与技术系,BNRist中心,THU-Bosch机器学习中心,清华大学);ShengShu(盛书);Gaoling School of Artificial Intelligence, Renmin University of China(北京理工大学人工智能学院,中国人民大学);The University of Texas at Austin(德克萨斯大学奥斯汀分校);Princeton University(普林斯顿大学)
IterInv: Iterative Inversion for Pixel-Level T2I Models
University of Chinese Academy of Sciences(中国科学院大学);Institute of Optics and Electronics(光电技术研究所);Computer Vision Center(计算机视觉中心);Universitat Autonoma de Barcelona(巴塞罗那自治大学)
Do Vision Models Learn Physical Constraints or Rendering Shortcuts? A Counterfactual Benchmark for Grounded Physical Consistency
Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)(转化神经影像与数据科学三机构中心);Georgia State University(佐治亚州立大学);Georgia Institute of Technology(佐治亚理工学院);Emory University(埃默里大学)