21:15
2026-10-05
qlabs.sh
machine-learning
Dust: Pretraining Transformers Without Backpropagation
Researchers Samip Dahal, Bishwas Mandal, Serdar Gülbahar and Akshay Vegesna released Dust, a zeroth-order pretraining method that perturbs activations independently at every token so one forward pass …