Hi there, I'm Yue ZHAO (赵越 in Chinese)! 👋
Note on External Advisory/Consultancy:
Dr. Zhao occasionally provides technical advice to selected projects on topics such as privacy-preserving AI and secure machine learning systems.
These collaborations are strictly technical in nature, with no involvement in financial operations, external fundraising, or investment-related activities.
😄 I am an Assistant Professor at USC Computer Science; see the latest information at my homepage.
My research centers on building reliable, safe, and scalable AI systems, with a focus on understanding and mitigating failure modes in modern foundation models and agentic systems.
I organize my work into two tightly connected tiers:
- Tier 1: advancing the scientific foundations of reliability and safety in modern AI systems
- Tier 2: translating these foundations into system-level evaluation frameworks and high-impact scientific and societal applications
I study why and how modern AI systems fail under distribution shift, uncertainty, and strategic pressure, and develop methods to make their behavior more predictable and reliable.
This tier comprises two complementary directions:
-
LLM & Agent Safety
Understanding and mitigating failure modes in large language models and agentic systems, including hallucinations, jailbreaks, privacy leakage, model extraction, and multi-agent instability. -
Robustness & Failure Detection
Developing algorithms and benchmarks to identify abnormal or unreliable behavior, grounded in robustness, out-of-distribution generalization, and anomaly detection.
Keywords:
LLM Safety, Robustness, Agents, Hallucination Mitigation, Jailbreak Detection, OOD Generalization, Failure Analysis
I adopt a system-oriented perspective to evaluate, stress-test, and deploy reliable AI in realistic settings, and apply these methods to domains where failures carry high cost.
This tier focuses on two areas that operationalize foundational advances:
-
Evaluation & Benchmarking
Designing scalable evaluation frameworks, benchmarks, and workflows that probe model and agent behavior under realistic and adversarial conditions. -
AI for Science & Society
Applying reliable foundation models to high-impact domains, including climate and weather forecasting, healthcare and biomedicine, and political or social decision-making.
Keywords:
Evaluation, Benchmarking, System-Level Analysis, AI for Science, Scientific Foundation Models, Climate & Weather Modeling, AI for Healthcare
📫 Contact me by: