Search Results for author: Weichen Fan

Found 8 papers, 5 papers with code

CFG-Zero*: Improved Classifier-Free Guidance for Flow Matching Models

1 code implementation24 Mar 2025 Weichen Fan, Amber Yijia Zheng, Raymond A. Yeh, Ziwei Liu

Classifier-Free Guidance (CFG) is a widely adopted technique in diffusion/flow models to improve image fidelity and controllability.

RepVideo: Rethinking Cross-Layer Representation for Video Generation

no code implementations15 Jan 2025 Chenyang Si, Weichen Fan, Zhengyao Lv, Ziqi Huang, Yu Qiao, Ziwei Liu

Video generation has achieved remarkable progress with the introduction of diffusion models, which have significantly improved the quality of generated videos.

Video Generation

Link-Context Learning for Multimodal LLMs

1 code implementation CVPR 2024 Yan Tai, Weichen Fan, Zhao Zhang, Feng Zhu, Rui Zhao, Ziwei Liu

The ability to learn from context with novel concepts, and deliver appropriate responses are essential in human conversations.

Few-Shot Learning In-Context Learning +1

Hierarchy Flow For High-Fidelity Image-to-Image Translation

1 code implementation14 Aug 2023 Weichen Fan, Jinghuan Chen, Ziwei Liu

In this work, we propose Hierarchy Flow, a novel flow-based model to achieve better content preservation during translation.

Image-to-Image Translation Style Transfer +1

StyleFlow For Content-Fixed Image to Image Translation

1 code implementation5 Jul 2022 Weichen Fan, Jinghuan Chen, Jiabin Ma, Jun Hou, Shuai Yi

We evaluate our model in several I2I translation benchmarks, and the results show that the proposed model has advantages over previous methods in both strongly constrained and normally constrained tasks.

Colorization Image-to-Image Translation +2

InvNorm: Domain Generalization for Object Detection in Gastrointestinal Endoscopy

no code implementations5 May 2022 Weichen Fan, Yuanbo Yang, Kunpeng Qiu, Shuo Wang, Yongxin Guo

Therefore, to address the generalization problem in GI(Gastrointestinal) endoscopy, we propose a multi-domain GI dataset and a light, plug-in block called InvNorm(Invertible Normalization), which could achieve a better generalization performance in any structure.

Domain Generalization Ethics +3

Cannot find the paper you are looking for? You can Submit a new open access paper.