Yasir Zaki

Associate Professor at New York University Abu Dhabi working on security, privacy, and bias in machine learning systems, particularly language models. Research focuses on how post-training procedures shape model behavior and social biases.

Sources in this wiki

• From Leniency to Suspicion: How an Open Post-Training Recipe Changes Whom LLM Fact-Checkers Believe

Topics

• Speaker Bias
• Fact-checking systems
• Bias in Language Models