Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Do Orthogonality Constraints Help Language-Model Training Beyond Sphere Optimization?
Kai Hu, Li Zhang, Weichen Yu +2
A growing body of work reports benefits from orthogonality constraints in fine-tuning large language models (LLMs). These results raise an open question: do orthogonality constrain…
cs.LG2026
LipNeXt: Scaling up Lipschitz-based Certified Robustness to Billion-parameter Models
Kai Hu, Haoqi Hu, Matt Fredrikson
Lipschitz-based certification offers efficient, deterministic robustness guarantees but has struggled to scale in model size, training efficiency, and ImageNet performance. We intr…