Rotem’s research explores whether large language models can capture aspects of human episodic memory. Her current research examines memory-augmented language models as computational models of human memory, comparing their predictions with neural activity recorded while people listen to naturalistic stories and later recall them.
Before joining the lab, Rotem completed her B.Sc. in Computer Science and Cognitive and Brain Sciences and her Ph.D. in Cognitive Neuroscience at the Hebrew University of Jerusalem. During her Ph.D., she combined behavioral experiments, computational modeling, and fMRI to study how the brain organizes autobiographical memories across timescales and how personally meaningful memories are represented in the brain. She also explored how large language models can be used to model aspects of human cognition, including personality.