Started AAAI 2027-oriented research on multimodal phishing website detection, supporting literature review, novelty analysis, experiment design, model fine-tuning, and experimental validation.
Research Focus
I am researching
I build research prototypes around AI safety, LLM agents, software security analysis, and reliable multimodal systems.
Profile
About Me
I am a junior undergraduate majoring in Computer Science and Technology at the University of Shanghai for Science and Technology. My current work centers on safety-aware LLM systems, runtime governance for agents, CodeQL-assisted security analysis, and efficient neural models for visual correspondence pruning.
I keep this public homepage focused on research, projects, open-source activity, and reproducible artifacts.
Contact and Representative Links
Open to academic discussion and collaboration
Please feel free to reach out by email for academic discussion, research collaboration, open-source projects, or learning exchange.
Recent updates
News
Joined Prof. Ma's team on an enterprise-oriented multi-agent secure collaboration project, contributing to requirements analysis, system design, implementation, and testing.
Completed an International Organization Internship through the UNDP-administered UNV platform on mangrove responses to climate change and produced a review manuscript on blue carbon mitigation.
Joined the XSafeClaw agent safety project and worked on runtime integration, controlled tool calls, and safety audit workflows.
Started SafeCodeRL, a multi-agent reinforcement learning framework for safety-constrained LLM code generation.
Worked on VulnSeeker, combining CodeQL static analysis with LLM-based security judgement.
Started SFMambaNet research on spectral-frequency enhanced state space models for correspondence pruning.
Timeline
Selected Research and Experience
Apr 2026 - Present
Research Intern, XSafeClaw Agent Runtime Safety
Fudan Trusted Embodied AI Institute
- Worked on XSafeClaw, an open-source safety platform for local and multi-runtime agent systems.
- Implemented runtime integration for nanobot-style local agent sessions, including session discovery, controlled tool calls, and unified platform display.
- Contributed to risk control workflows, validation, and documentation for explainable agent governance.
- The public XSafeClaw repository had 150 GitHub stars when this site plan was prepared.
Jan 2026 - Mar 2026
Research Intern, VulnSeeker
Fudan Software Engineering Lab
- Built an automated code security analysis workflow that combines CodeQL static analysis with LLM-based vulnerability judgement.
- Designed prompt workflows for CodeQL and LLM interaction, including structured status codes for security decisions.
- Implemented code context expansion, vulnerability detection after code generation, and combined rule-based plus model-based assessment.
Jan 31, 2026 - May 21, 2026
International Organization Internship, UNDP-administered UNV Online Research Intern
Morobe Development Foundation Inc., Papua New Guinea
- Completed an International Organization Internship through the UNDP-administered United Nations Volunteers (UNV) platform, supporting research on mangrove responses to climate change scenarios and blue carbon mitigation.
- The United Nations Volunteers (UNV) programme is administered by the United Nations Development Programme (UNDP).
- Synthesized literature on sea-level rise, coastal squeeze, nutrient loading, carbon sequestration, and national climate policy.
- Produced a review manuscript titled Research Support in Analysing Mangrove Responses to Climate Change: The Interplay of Anthropogenic Impacts and Blue Carbon Mitigation.
Feb 2026 - Present
First-author Research, SafeCodeRL
Independent research collaboration
- Proposed SafeCodeRL, a multi-agent framework for dynamic safety constraints in LLM code generation.
- Designed a closed-loop collaboration workflow across five agents and a PPO-style constraint-aware policy.
- Reported a large reduction in high-risk vulnerable code generation while preserving functional correctness in the manuscript.
Sep 2025 - Feb 2026
First-author Research, SFMambaNet
Multimedia Intelligent Computing and Security Lab
- Proposed SFMambaNet, a spectral-frequency enhanced selective state space model for correspondence pruning.
- Explored frequency-aware global context modeling as an efficient alternative to quadratic-complexity Transformer designs.
- Completed literature review, core idea design, experiments, and manuscript writing under faculty supervision.
Research output
Selected Publications and Manuscripts
SafeCodeRL: A Multi-Agent Reinforcement Learning Framework for Safety-Constrained LLM Code Generation
First author · Published on June 2, 2026 · 2026
SFMambaNet: Spectral-Frequency Enhanced Selective State Space Model for Correspondence Pruning
First author · arXiv preprint and under review · 2026
Research Support in Analysing Mangrove Responses to Climate Change: The Interplay of Anthropogenic Impacts and Blue Carbon Mitigation
Author · review manuscript · 2026
Live Open Source Signal
GitHub Commit Activity
Last 12 months
Loading public contribution graph...Syncing
Live contributions are fetched from a public GitHub contribution API and may be cached.
Builds and artifacts
Featured Projects
2026 Featured
International Organization Internship: UNDP-administered UNV Mangrove Research
A UNDP-administered UNV research internship output on sea-level rise, nutrient loading, coastal squeeze, and blue carbon mitigation.
2026 Featured
XSafeClaw: Agent Runtime Safety Governance
An open-source agent safety platform for runtime session discovery, controlled tool calls, human approval, and risk auditing.
2026 Featured
SafeCodeRL: Safety-Constrained LLM Code Generation
A multi-agent reinforcement learning framework for dynamic security constraints in LLM code generation.
2026 Featured
VulnSeeker: CodeQL and LLM Security Analysis
An automated security analysis pipeline that combines CodeQL static analysis and LLM-based vulnerability judgement.
Knowledge progress
Learning Map
AI Safety and LLM Agents
88%Focused on agent runtime governance, safety-constrained code generation, tool-call auditing, and human-in-the-loop risk control.
Software Security and Program Analysis
84%Built workflows around CodeQL, LLM-assisted vulnerability judgement, code context expansion, and post-generation security inspection.
Deep Learning and Multimodal Intelligence
82%Studied neural architectures for visual correspondence, spectral-frequency modeling, state space models, and robust image watermarking.
Algorithms and Systems Programming
86%Maintains strong C/C++ foundations through algorithmic problem solving, contest-style implementation, and performance-conscious coding.
Scientific Writing and Research Methodology
80%Practices literature synthesis, experiment design, manuscript organization, and cross-domain research communication.
Web Engineering and Research Prototyping
78%Uses TypeScript, Python, Git, LaTeX, and static-site tooling to turn research ideas into reproducible public artifacts.
Signals
Awards and Skills
Selected Honors
- National First Prize, RAICOM Robot Developer Competition, 2025
- National Second Prize, LanQiao Cup C/C++ Programming Contest, 2025
- Silver Award, CCPC Shanghai Collegiate Programming Contest, 2025
- National Third Prize, National English Competition for College Students, 2025
- Shanghai First Prize, National College Student Mathematics Competition, 2024
- First-author software copyright for a robust deep-learning watermark mobile application