AIKeivan Rezaei defends PhD on editing and unlearning in foundation models
His adviser, Soheil Feizi, congratulated him and praised his research on foundation models, model editing, unlearning and robustness.
TLDR
Keivan Rezaei announced on September 18, 2026, that he had successfully defended his PhD dissertation, “Localize, Edit, Unlearn: Structural Interventions and Data-Level Control in Foundation Models.” His adviser, Soheil Feizi, praised his research and described watching a PhD student graduate as a “happy and bittersweet moment.”
Combined views
4.8K
1 Source, first seen 1d ago
likes
Keivan Rezaei defends PhD on editing and unlearning in foundation models
His adviser, Soheil Feizi, congratulated him and praised his research on foundation models, model editing, unlearning and robustness.