5 papers
Learning to Be Guided: Shaping Latent Geometry for Classifier-Free Guidance
Haoming Liu, Junhao Zhu, Shaoyi Zheng +4
Classifier-free guidance (CFG) is a standard component of conditional diffusion models, and guided generation performance has become the \textit{de facto} measure of state-of-the-a…
Growth-induced trade-off between polarity, electrical isolation, and structural quality in AlN grown on Si(111) by plasma-assisted molecular beam epitaxy
Hossein Yazdani, Nhu Tran, Xin Du +3
We investigate the effects of substrate temperature and Al flux, at a fixed N flux, on the polarity, electrical properties, and structural quality of AlN films grown on high-resist…
TFP: Temporally Conditioned Memory-Fusion Policies for Visuomotor Learning
Yushen Liang, Yue Peng, Baosheng Jin +6
Vision--Language--Action (VLA) policies such as and OpenVLA perform well on many manipulation tasks, but they are often reactive: the next action is predicted from the cu…
Histogram-constrained Image Generation
Haoming Liu, Yuanhe Guo, Yijia Cao +2
Diffusion models have emerged as a dominant paradigm in generative modeling, enabling high-fidelity sampling from complex data distributions. Despite impressive capabilities, contr…
From Navigation to Refinement: Revealing the Two-Stage Nature of Flow-based Diffusion Models through Oracle Velocity
Haoming Liu, Jinnuo Liu, Yanhao Li +5
Flow-based diffusion models have emerged as a leading paradigm for training generative models across images and videos. However, their memorization-generalization behavior remains…