Haomin Zhuang
Scholar

Haomin Zhuang

Google Scholar ID: vXllNroAAAAJ
University of Notre Dame
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
228
 
H-index
6
 
i10-index
5
 
Publications
15
 
Co-authors
0
 
Contact
Emailhzhuang2@nd.eduGitHubOpen ↗
Publications
20 items
MemoHarness: Agent Harnesses That Learn from Experience
2026
Cited
0
ASAP: Agent-System Co-Design for Wall-Clock-Centered Auto HPO Research for ML Experiments
2026
Cited
0
Getting Better at Working With You: Compiling User Corrections into Runtime Enforcement for Coding Agents
2026
Cited
0
AIRGuard: Guarding Agent Actions with Runtime Authority Control
2026
Cited
0
AgentTrap: Measuring Runtime Trust Failures in Third-Party Agent Skills
2026
Cited
0
SkillGen: Verified Inference-Time Agent Skill Synthesis
2026
Cited
0
AgentClick: A Skill-Based Human-in-the-Loop Review Layer for Terminal AI Agents
2026
Cited
0
Guardian-as-an-Advisor: Advancing Next-Generation Guardian Models for Trustworthy LLMs
2026
Cited
0
Resume
Academic Achievements
  • Reviewer for ICLR 2025, ACL 2025, and TDSC.
Research Experience
  • Research intern in OPTML group, advised by Prof. Sijia Liu; research intern in IntelliSys Lab, guided by Prof. Hao Wang.
Education
  • Ph.D. student in MINE Lab, advised by Prof. Xiangliang Zhang. Bachelor's degree in Software Engineering from South China University of Technology.
Background
  • Research interests include large language models (LLMs), AI safety, and advancing reasoning capabilities in LLMs. Focused on model substructures and embeddings, aiming to improve their efficiency, interpretability, and alignment with human reasoning.