Scalable Oversight
Supervising capable but unreliable experts on complex tasks using multiagent debate.
Salman Rahman
I am a Computer Science graduate student at New York University. My research goal is to improve the trustworthiness of artificial intelligence systems by ensuring AI systems smarter than humans align with human interests and evaluating and mitigating the catastrophic risks posed by AI systems. My research takes shape in two key directions:
Model Evaluation for Extreme Risks
Evaluation and mitigation of risks posed by large models.
Earlier in my academic journey, I delved into computational sustainability during my undergrad and master's studies. I was captivated by the prospect of leveraging technology to provide solutions to the pressing societal and environmental challenges of the 21st century. Explore my publications on computational sustainability here.
News
Mar 30, 2024
One paper on machine learning explanation disparity is accepted to ACM FAccT 2024.
Selected Publications
Generalization in Healthcare AI: Evaluation of a Clinical Large Language Model
arXiv preprint, 2024 PDF
Teaching
Fall 2023
Guest Lecturer, Foundation(Large) Language Model
CS-GY 9223: Foundations of Data Science, Graduate, NYU