I'm a PhD student in Electronic Information at Fudan University, specializing in Safety & Privacy, Large Language Models Safety, and Agent Safety research. My work focuses on bridging theoretical research with practical applications in artificial intelligence and cybersecurity.
- Current Research: LLM Security, Endogenous Security, AI-enabled Traditional Security
- Exploring: Safety & Privacy, Large Language Models Safety, Agent Safety
- Education: PhD @ Fudan University (2025-Present) | B.S. @ Shanghai Maritime University
- Visiting Student: HKUST
| Project | Description | Tech |
|---|---|---|
| TIPExploit | Empirical risk assessment of tool-invocation prompts in LLM-based agentic systems (Under Review) |  |
| Project | Description | Tech |
|---|---|---|
| Precious-Metals-Investment-Analysis | Quantitative analysis and investment strategy for precious metals market | |
| SoulByte | Transform WeChat chat records into AI training datasets and personal knowledge bases | |
| NexusAI-Hub | Unified platform for managing multi-provider AI models with OpenAI-compatible APIs |  |
| mcp_based_subsystem | MCP-based subsystem implementation for AI agent coordination | |
| Project | Description | Tech |
|---|---|---|
| syzkaller (fork) | Enhanced kernel fuzzer with LLM-assisted mutation for improved coverage |  |
LLM Security ████████████████████░ 95%
Endogenous Security ██████████████████░░░ 85%
Deep Learning █████████████████░░░░ 80%
Kernel Fuzzing ███████████████░░░░░░ 75%
Knowledge Graphs ██████████████░░░░░░░ 70%
- Open to collaborate on: LLM security, endogenous security, AI-driven vulnerability detection
- Ask me about: LLM security challenges, AI applications in security domains, or any of my research projects!