works on
attention mechanisms 1attention reranking 1language model sampling 1linear-time models 1long-context recall 1masked diffusion 1parallel decoding 1sequence modeling 1sparse memory 1training-free methods 1

From the 2 of 58 linked papers with an AI index.

activity
20242026
collaborators
Showing math.OCShow all

1 paper · 1 filter

math.OC2026

Constrained Stochastic Spectral Preconditioning Converges for Nonconvex Objectives

Konstantinos Oikonomidis, Jan Quan, Kimon Antonakopoulos +3

In this work, we develop proximal preconditioned gradient methods with a focus on spectral gradient methods providing a proximal extension to the Muon and Scion optimizers. We intr…