About Me

I’m David, a student in Data Science and Big Data Technology at Southern University of Science and Technology, Shenzhen. I focus on building AI agents and turning complex workflows into clear, polished and efficient products. I enjoy making tools that reduce repetitive work and help people get things done.

Education

2023 - 2027

Southern University of Science and Technology emblem

Southern University of Science and Technology

Bachelor’s degree in Data Science and Big Data Technology

Research Interests

  • Large Language Models and AI Agents
  • Agent Workflows and Tool Use
  • Group-Robust Learning

Publications

The two-stage ProME training, classifier-repair and deployment framework from the paper

ProME: Prototype-Margin Environments with Repair-Aware Selection for Group-Robust Learning

Qianqian Wang, Yunshan Li, Dawei Huang, Wenwu Gong, Lili Yang

arXiv
Figure 2 from GRA-Flow: image-space correction decomposition under low-light and spatial haze, including channel-mean and centered components and pixelwise error changes.

GRA-Flow: Gain-Risk Aligned Rectified-Flow Refinement for Task-Free OOD Image Restoration

Qianqian Wang, Dawei Huang, Hang Fu, Wenwu Gong, Yunshan Li, Lili Yang

Figure 1 from RECAST: residual-transport training, full- and half-resolution inference, scale-consistency fusion and view averaging on a frozen backbone.

Controlled Residual Transport for Frozen-Backbone All-in-One Image Restoration

Qianqian Wang, Wenwu Gong, Dawei Huang, Hang Fu, Yunshan Li, Lili Yang

Projects

Evidence Tracker workflow: paper identity checks, full-text evidence retrieval, claim review and an auditable report

Evidence Tracker

Research evidence plugin

Code