Zhijie Zheng (郑志杰)

I am a junior undergraduate student at Beihang University (BUAA), supervised by Prof. Lu Sheng. I am currently a research intern at Shanghai AI Lab, advised by Prof. Jing Shao.

My research focuses on incorporating human priors into AI systems to make them more reliable and efficient. I am particularly interested in reinforcement learning and multimodal reasoning.

Email  /  Google Scholar  /  Github

News

• [2026.02]   🎉 One paper accepted by CVPR 2026!
• [2026.01]   🎉 One paper accepted by ICLR 2026!

Selected Publications

(*, †, ‡ indicates equal contributions, corresponding author, and co-leads, respectively.)

A 4B step-level guard for pre-execution safety checks and trajectory auditing, trained with StepGen and Balance-GRPO.

First benchmark modeling the full lifecycle of multi-agent financial fraud and collusion on social platforms.

A general-purpose, training-free agentic framework for spatial reasoning that bridges the semantic-to-geometric gap in VLMs.

Technical Reports (Teamwork)

A lightweight and scalable alignment framework for AI agent safety, with taxonomy-guided data synthesis and efficient SFT/RL post-training.

Comprehensive risk assessment of frontier AI models across seven critical areas including cyber offense, biological risks, and autonomous AI R&D.

Education

• 2023.09 - 2027.06 (expected), B.Eng., Beihang University, Beijing, China.

Experience

• 2025.04 - present, Research Intern, Shanghai AI Lab. Advisor: Prof. Jing Shao.

Template from Jon Barron.