Jiazhe Wei

Jiazhe Wei

Biography

Hi! I am Jiazhe Wei (魏佳哲), an incoming master student majoring in Artificial Intelligence at Nanjing University and doing research at Pattern Recognition Lab @NJU, supervised by Prof. Chenyang Si and Prof. Tieniu Tan.

My research interests mainly focus on: 1) Highly Controllable Visual Generative Models; 2) World Foundation Models; 3) Efficient Generative Models and Model Compression; 4) Unified Multi-modal Learning.

✨ I’m always open to all kinds of cooperation and discussion. You can contact me via email or WeChat: wjz_void.

🔥 News

  • 2025.12:  🎉 PosterCopilot is now available on arxiv, code is under construction.

  • 2025.09:  🏆 Awarded CCF Elite Collegiate Award of 2025. See you in Harbin.

  • 2025.07:  🎉 ALM-PU is accepted by ECML-PKDD 2025 as oral presentation. Thanks to all co-authors. See you in Porto, Spain.

  • 2025.02:  🎉 NI-GDBA is accepted by WWW 2025 as oral presentation. Congratulations to Ken Li! See you in Sydney, Australia.

📝 Publications | Preprints Full List

Arxiv
PosterCopilot

PosterCopilot: Toward Layout Reasoning and Controllable Editing for Professional Graphic Design

Jiazhe Wei *, Ken Li *, Tianyu Lao, Haofan Wang, Liang Wang, Caifeng Shan, Chenyang Si

[PDF] [Project Page] [Github]

TL;DR: We introduce PosterCopilot, the first graphic design framework that fully integrates visual feedback with human aesthetics. Leveraging our Generative Agent and the design model’s powerful reasoning capabilities, it enables layer-level, multi-round, high-fidelity editing.

ECML-PKDD 2025 Oral
ALM-PU

ALM-PU: Positive and Unlabeled learning with constrained optimization

Jiazhe Wei*, Yuefei Wu*, Ken Li, Bo Dong, Bin Shi†

[PDF] [Github]

TL;DR: We propose a novel constraint optimization-based PU-learning framework, which enhances sample discrimination capability by imposing distinct mathematical constraints on samples and resolving the problem through systematic optimization.

WWW 2025 Oral
NI-GDBA

NI-GDBA: Non-Intrusive Distributed Backdoor Attack Based on Adaptive Perturbation on Federated Graph Learning

Ken Li, Bin Shi, Jiazhe Wei, Bo Dong

[PDF]

TL;DR: We propose a non-intrusive graph distributed backdoor attack(NI-GDBA) that does not require backdoor triggers to be injected in the training data.

* Equal contribution. † Corresponding author.

🔬 Research & Intern Experience

lovart

Lovart, AIGC Research Team

2025.08 - Present, Intern Research Scientist, Shanghai, China

Research Topics: 1) Interactive Image Editing, 2) Efficient Generative Models, 3) Multi-modal learning

Working with Haofan Wang.

LinsLab

LinsLab, Learning and Inference Systems Lab, Westlake University & Zhejiang University

2024.11 - 2025.02, Visiting Student, Hangzhou, China

Research Topics: Efficient Diffusion Models and Flow Matching Models

Working with Prof. Tao Lin and Dr. Peng Sun

CERN

European Organization for Nuclear Research & CERN

2025.01 - 2025.02, Visiting Student, Geneva, Switzerland

Supervised by Prof. Zhili Weng

BDKElab

BDKElab, Xi'an Jiaotong University

2023.09 - 2024.10, Research Assistant, Xi'an, China

This was a period filled with bitterness and twists and turns -- a detour I had to take in my research career. Despite this, I remain grateful for the experience, as it taught me early on what kind of taste an independent researcher should cultivate, what abilities they must possess, and the importance of reaching outward to embrace the whole world.

🎖 Honors and Awards (Selected)

  • 2025 CCF Elite Collegiate Award, Awarded to 99 top Chinese undergraduates major in CS&AI nationwide
  • 2024 Baosteel Outstanding Student Scholarship, Awarded to 400 top Chinese undergraduates&graduates nationwide
  • 2024 350 points, ranked in top 2% nationwide, CCF Certified Software Professional(CSP)
  • 2024 National Third Prize 🥉, Blue Bridge Cup National Software and Information Technology Professionals Competition
  • 2023 Gold Medal 🥇, Bai Du–A star Programming Competition
  • 2023 First scholarship for Outstanding Students
  • 2023, 2024, 2025 Dean’s list
  • 2023 Silver Medal, (CCPC) China Collegiate Programming Contest National Invitational Contest
  • 2023 HuaWei scholarship
  • 2023 National third prize, Chinese College Students English Academic Proficiency Competition

📖 Education

XJTU

Xi'an Jiaotong University (XJTU), China

2022.09 - 2026.06 (expected)

B.Eng. in Computer Science and Technology (Qian Xuesen Honor Class)

• GPA: 91.97 / 100

• Ranking: 3/43

💬 Talks ＆ Reports

📋 Service

  • Reviewer: AAAI 2025, ECML-PKDD 2025
  • Co-Founder of XJTU Deep Learning Seminar

Latest updated in Dec. 2025

© Jiazhe Wei