Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
PowerStep: Memory-Efficient Adaptive Optimization via -Norm Steepest Descent
Yao Lu, Dengdong Fan, Shixun Zhang +1
Adaptive optimizers such as Adam are standard for training Transformers, but storing gradient first and second moments incurs substantial memory overhead. We introduce PowerStep, a…
cs.LG2026
Sub-JEPA: Subspace Gaussian Regularization for Stable End-to-End World Models
Kai Zhao, Dongliang Nie, Yuchen Lin +4
Joint-Embedding Predictive Architectures (JEPAs) provide a simpleframework for learning world models by predicting future latent representations.However, JEPA training is subject t…