Simran Arora

Simran Arora is a researcher at the California Institute of Technology, Department of CS. Her research in AI systems focuses on expanding the Pareto frontier between quality and efficiency, to unlock new AI capabilities. Arora's work explores methods to enhance AI performance while maintaining computational efficiency. Her research has the potential to impact various domains by enabling more effective and scalable AI solutions. Arora maintains a professional website at https://arorasimran.com/.

Simran Arora is a computer science researcher at Stanford University specializing in AI systems. Her work explores the opportunities and risks of foundation models, focusing on trustworthiness, data wrangling, and efficient language models. She investigates methods to improve recall, throughput, and structured data generation using language models. Her research includes benchmarking long-context retrieval models, enhancing self-consistency in large language models, and optimizing graph queries. Arora also explores AI systems for GPU kernels and molecular sensors, emphasizing practical applications and performance. Her contributions span both theoretical and applied aspects of AI, with an emphasis on scalability and reliability.