Multi-subject video generation faces two key challenges: uncontrollable fidelity strength and potential semantic drift. We address these by analyzing the internal mechanisms of Diffusion Transformers (DiTs). We found that certain attention blocks naturally form an Intrinsic Spatial Grounding Map (ISGM) that precisely l...
N. Yu, Ye Tian, Biao-Long Chen et al.· 0 citations
Virtual try-on (VTON) requires precise pixel-level fidelity, yet mainstream Diffusion Transformers (DiTs) often suffer from texture degradation and structural drift. We identify symmetric interactions in standard joint-attention mechanisms as a source of these failures. Although such interactions support semantic flexi...
Zi-Shu Qin, Zhi-Yu Jin, Pi-Pei Huang et al.· 0 citations
InfinityEdit is proposed, a lightweight edit adapter that equips a streaming video generator with unbounded editing ability that faithfully continues the stream under each edit, and stays stable over unbounded edit sequences.
Yunze Tong, Mu-Shui Liu, Can-Yu Zhao et al.· 0 citations
This work formally identifies this limitation as High-Frequency Trajectory Collapse: supervised fine-tuning converges to the conditional mean of the training distribution, which is dominated by smooth, low-frequency textures, causing high-frequency patterns to become nearly un-sampleable.
Yuhan Li, Xianfeng Tan, Fan-Gao Zeng et al.· 0 citations
Self-OPD is introduced, a teacher-free OPD framework for flow matching models that turns the student's own self-exploration into step-wise supervision and outperforms prior RL and OPD methods without task-specific teachers.
Shi-Yi Zhang, Mu-Shui Liu, Yunze Tong et al.· 1 citation
This work proposes REST (Reward-Enhanced Scored-Trajectory Distillation), a single-stage RL-distillation co-training framework that attaches a decoupled student to an arbitrary RL teacher that enables few-step CFG-free inference that matches or surpasses its 40-step RL teacher, with an overall additional training cost...
Yuhan Li, Fan-Gao Zeng, Sicong Kang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
