Revisit Large-Scale Image-Caption Data in Pre-training Multimodal Foundation Modelsby Zhengfeng Lai, Vasileios Saveris, Chen Chen,…
GUD: Generation with Unified Diffusionby Mathis Gerdes, Max Welling, Miranda C. N. ChengFirst submitted to…
Extracting Training Data from Unconditional Diffusion Modelsby Yunhao Chen, Shujie Wang, Difan Zou, Xingjun MaFirst…
SAFLEX: Self-Adaptive Augmentation via Feature Label Extrapolationby Mucong Ding, Bang An, Yuancheng Xu, Anirudh Satheesh,…
Local Flow Matching Generative Modelsby Chen Xu, Xiuyuan Cheng, Yao XieFirst submitted to arxiv on:…
Convergence of Score-Based Discrete Diffusion Models: A Discrete-Time Analysisby Zikun Zhang, Zixiang Chen, Quanquan GuFirst…
Simplicity bias and optimization threshold in two-layer ReLU networksby Etienne Boursier, Nicolas FlammarionFirst submitted to…
Eliminating Oversaturation and Artifacts of High Guidance Scales in Diffusion Modelsby Seyedmorteza Sadat, Otmar Hilliges,…
Channel-aware Contrastive Conditional Diffusion for Multivariate Probabilistic Time Series Forecastingby Siyang Li, Yize Chen, Hui…
EC-DIT: Scaling Diffusion Transformers with Adaptive Expert-Choice Routingby Haotian Sun, Tao Lei, Bowen Zhang, Yanghao…