AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Understanding and improving Shampoo and SOAP via Kullback-Leibler Minimization

Wu Lin, Scott C. Lowe, Felix Dangel, Runa Eschenhagen, Zikun Xu, Roger Baker Grosse

ICLR 2026 · Architectures and efficiency › RNNs and normalization · approach: Scalable approximation and equivariant compute · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use KL-Shampoo's EMA eigenvalue estimation with QR eigenbasis instead of step-size grafting with AdamKL-Shampoo · Language model pre-trainingvalidation lossImproved
replace instantaneous eigenvalue estimation with the EMA scheme (ablation)Shampoo variants · Language model pre-trainingvalidation lossImproved
replace Frobenius norm or von Neumann divergence with KL divergence (ablation)Shampoo schemes · Language model pre-trainingvalidation lossImproved
add KL-based two-sided estimation to SOAP (KL-SOAP) instead of Adam in eigenbasisSOAP · Language model pre-trainingvalidation lossImproved
use KL-Shampoo instead of KL-SOAPKL-SOAP · Language model pre-trainingvalidation lossImproved
apply EMA eigenvalue scheme to trace-scaling (VN) Shampoo (ablation)VN-Shampoo · Language model pre-trainingvalidation lossImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy