Our paper When is Task Vector Provably Effective Model Editing? A Generalization Analysis of Nonlinear Transformers is accepted to ICLR 2025 as an Oral Presentation (only 1.8% acceptance rate)!
Our paper When is Task Vector Provably Effective Model Editing? A Generalization Analysis of Nonlinear Transformers is accepted to ICLR 2025 as an Oral Presentation (only 1.8% acceptance rate)!