Project Page Highlights

Projects – Furong Huang

Projects

Explore a selection of projects from our research group, categorized by themes such as security, fairness, and generalization.

Ensuring robustness and fairness in machine learning under worst-case adversarial conditions.

Analyzing vulnerabilities in vision-language models through poisoning attacks.

Testing false refusals in large language models to ensure robustness.

Generating adversarial backdoor models to enhance AI security.

Addressing long-term fairness in sequential decision-making processes.

Evaluating the robustness of invisible watermarks under adversarial attacks.

Benchmarking LLM performance across standardized difficulty levels.

Building modular AI agents for embodied intelligence.