Ninareh Mehrabi

Ninareh Mehrabi is a research scientist at Amazon AGI Foundations working on AI safety, red teaming, and robustness of generative models. Her research focuses on identifying and mitigating vulnerabilities in large language models and multimodal AI systems.

Sources in this wiki

• FLIRT: Feedback Loop In-context Red Teaming

Topics

• Red teaming, Model safety, AI Safety

Notes

Lead author of FLIRT, an automated red teaming framework for evaluating generative models.