9 papers
Awakening Diffusion Transformers: Eliciting Stronger Generation and Understanding via Massive Activation Modulation
Chaofan Gan, Zicheng Zhao, Yuanpeng Tu +6
Massive Activations (MAs) have been widely observed in Transformer-based models, yet their structure and functional roles in Diffusion Transformers (DiTs) remain insufficiently und…
Bridging the Embodiment Gap: Disentangled Cross-Embodiment Video Editing
Zhiyuan Li, Wenyan Yang, Wenshuai Zhao +4
Learning robotic manipulation from human videos is a promising solution to the data bottleneck in robotics, but the distribution shift between humans and robots remains a critical…
MegaStyle: Constructing Diverse and Scalable Style Dataset via Consistent Text-to-Image Style Mapping
Junyao Gao, Sibo Liu, Jiaxing Li +6
In this paper, we introduce MegaStyle, a novel and scalable data curation pipeline that constructs an intra-style consistent, inter-style diverse and high-quality style dataset. We…
PlayerOne: Egocentric World Simulator
Yuanpeng Tu, Hao Luo, Xi Chen +3
We introduce PlayerOne, the first egocentric realistic world simulator, facilitating immersive and unrestricted exploration within vividly dynamic environments. Given an egocentric…
Unleashing Diffusion Transformers for Visual Correspondence by Modulating Massive Activations
Chaofan Gan, Yuanpeng Tu, Xi Chen +4
Pre-trained stable diffusion models (SD) have shown great advances in visual correspondence. In this paper, we investigate the capabilities of Diffusion Transformers (DiTs) for acc…
Massive Activations are the Key to Local Detail Synthesis in Diffusion Transformers
Chaofan Gan, Zicheng Zhao, Yuanpeng Tu +5
Diffusion Transformers (DiTs) have recently emerged as a powerful backbone for visual generation. Recent observations reveal \emph{Massive Activations} (MAs) in their internal feat…