Showing 2026 · cs.LGShow all
2 papers · 2 filters
cs.LG2026
Training Dynamics of Softmax Self-Attention: Fast Global Convergence via Preconditioning
Gautam Goel, Mahdi Soltanolkotabi, Peter Bartlett
We study the training dynamics of gradient descent in a multilayer softmax self-attention network trained to perform linear regression and show that in the infinite-data limit this…
cs.LG2026
CrispEdit: Low-Curvature Projections for Scalable Non-Destructive LLM Editing
Zarif Ikram, Arad Firouzkouhi, Stephen Tu +2
A central challenge in large language model (LLM) editing is capability preservation: methods that successfully change targeted behavior can quietly game the editing proxy and corr…