SOURCE-LINKED INTELLIGENCE

Beyond the Matrix Sign: Quadratic Spectral Descent

arXiv · AI, language, vision and robotics · article · Sep 7, 2026 · UTC

Muon can be interpreted as optimizing a linear local objective over a spectral-norm ball. This gives a matrix-sign update that preserves the singular directions of the gradient and assigns the same magnitude to all active singular modes. We ask whether these two properties remain optimal when local curvature is taken into account. To answer this question, we keep Muon's spectral-norm constraint unchanged and replace the linear local model with a quadratic one. We call the resulting method \emph{Quadratic Spectral Descent} (QSD). We show that curvature can change both the singular values and th

Read original source ↗ Open in workspace

recordType

paper

region

Global