#machine learning
Apr 2026
MOONSHOT : A Framework for Multi-Objective Pruning of Vision and Large Language Models
This work proposes MOONSHOT, a general and flexible framework that extends any single-objective pruning method into a multi-objective formulation by jointly optimizing both the layer-wise reconstruction error and second-order Taylor approximation of the training loss.
Gabriel Afriat, Xiangui Meng, Shibal Ibrahim et al.
· Trans. Mach. Learn. Res. · 0 citations