Research topics include:

• AI safety for LLMs and multimodal models
• Safety alignment and robustness after model adaptation
• Unlearning, watermarking, and deepfake robustness
• Constitutional AI and scalable oversight

Research topics included:

• Safety of Multilingual LLMs
• Red Teaming of Large Language Models
• Llama Language Models

Research topics included:

• Computer vision algorithms for underwater robotics and pick-and-place systems