Qiucheng Wu

I am a 3rd year Ph.D. student in Computer Science at University of California, Santa Barbara, under supervision of Prof. Shiyu Chang. I received my bachelor and master's degrees from University of Michigan College of Engineering. I also receivied another bachelor degree from and Shanghai Jiao Tong University UM-SJTU Joint Institute.

My current research focuses on generative models in vision and language. Particularly, I am interested in controllable generation leveraging high-quality pretrained generative models, including diffusion models and GANs. Besides, my general research interests span natural language processing and trustworthy machine learning. Some of my projects and publications related to these fields are demonstrated below.

Email: qiucheng@ucsb.edu /  CV  / 

* indicates authors with equal contributions.

We propose a new text-to-image algorithm with explicit control over cross-attention in diffusion models from spatial and temporal views. This alleviates inconsistencies between images and text and helps to fix errors like missing objects, mismatched attributes, and mislocated objects.

Based on a fixed stable diffusion model, we disentangle target attributes from a single training image. The learned parameters can then be applied to unseen images and achieve same edits. This finding leads to a lightweight image editing framework with only 50 learnable parameters.

We propose Ada-Deblur, a super-network that can be applied to a "broad spectrum" of various blur levels with no re-training on novel blurs.

Meta reinforcement learning requires substantial amounts of data. To improve the sample efficiency and performance of metal-RL algorithms on sparse-reward tasks, we introduce a novel objective function to learn an action translator among training tasks.

We propose DataSifter-Text to protect privacy of sensitive textual dataset. The DataSifter-Text obfuscates identifiable information in the dataset. It effectively balances between privacy and utility.

We apply the compressive big data analytics (CBDA) to analyze sailent features and key biomarkers of a high-dimensional dataset.

Most Likely Health Impact in 4th Annual Symposium Poster Competition in Michigan Institute for Data Science, University of Michigan.

We propose the DataSifter to obfuscate sensitivie dataset while preserve the utility for other researches.

Most Interesting Methodological Advances in 4th Annual Symposium Poster Competition in Michigan Institute for Data Science, University of Michigan.

In this capstone project, we designed a framework to boost web communications between terminal devices and servers by introducing edge servers. We use kubernets to organize edge servers as different containers. The edge servers are responsible for computing and sending information to terminal devices, leading to more efficient performance.

This website is built using the source code from Jon Barron's public academic website.