muon-optimizer

Tag

Cards List
#muon-optimizer

LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes

Hugging Face Daily Papers · yesterday Cached

LLaDA-Image presents a unified framework that combines a 6B diffusion transformer with a frozen vision-language module for generating photorealistic images with precise editing, achieving state-of-the-art results among open-source models through efficient training and fast inference.

0 favorites 0 likes
#muon-optimizer

Muon with Finite Newton-Schulz: The Smoothing Benefit in Nonsmooth Nonconvex Optimization

arXiv cs.LG · 2026-08-28 Cached

This paper analyzes how finite Newton-Schulz iterations in the Muon optimizer benefit nonsmooth nonconvex optimization by smoothing the polar map, providing convergence guarantees that match best-known bounds.

0 favorites 0 likes
#muon-optimizer

Scaling Muon for Diffusion Transformers

arXiv cs.LG · 2026-08-24 Cached

This paper scales the Muon optimizer for Diffusion Transformers from 1.3B to 15B parameters, introducing Periodic Row-wise Muon to reduce computational overhead while preserving generative quality improvements over AdamW.

0 favorites 0 likes
#muon-optimizer

Federated Compositional Muon Optimizer for Matrix-Wise Models

arXiv cs.LG · 2026-08-14 Cached

This paper proposes FedCoMuon, a federated compositional Muon optimizer for matrix-wise models, along with a variance-reduced variant (FedCoMuon-VR). The authors provide convergence analysis under non-i.i.d. and non-convex settings, showing improved sample complexity over existing FedMuon algorithms, and demonstrate competitive performance on robust federated learning and task-distributed risk-sensitive meta learning.

0 favorites 0 likes
#muon-optimizer

The Active Ingredient in Muon's Grokking

arXiv cs.LG · 2026-07-24 Cached

This paper ablates the Muon optimizer to find that orthogonalization (Newton-Schulz iteration), not spectral scaling, is the key ingredient behind its faster grokking on modular arithmetic, and introduces a stability-aware metric for measuring grokking speed.

0 favorites 0 likes
#muon-optimizer

How Much Orthogonalization Does Muon Need?

arXiv cs.LG · 2026-06-02 Cached

This paper studies how much orthogonalization the Muon optimizer requires, proposing a five-step cubic Newton-Schulz schedule that reduces computational cost while achieving training quality similar to more expensive methods across GPT-2 Small and hybrid MoE/Mamba models.

0 favorites 0 likes
#muon-optimizer

@zhaoran_wang: for me, the coolest finding is that you can connect/interpolate all softmax/linear variants and give a promising direct…

X AI KOLs Timeline · 2026-05-30 Cached

Discussion of a finding that all softmax/linear attention variants can be interpolated, and that the Muon optimizer is crucial for Parallax to move beyond Softmax Attention. Includes link to paper and code.

0 favorites 0 likes
#muon-optimizer

SignMuon: Communication-Efficient Distributed Muon Optimization

arXiv cs.LG · 2026-05-19 Cached

SignMuon is a 1-bit, matrix-aware optimizer for distributed training that combines signSGD's majority-vote sign aggregation with Muon's polar-step framework, achieving 32x bandwidth reduction over float32 while maintaining strong convergence and performance on benchmarks like CIFAR-10/ResNet-50 and nanoGPT.

0 favorites 0 likes
← Back to home

Submit Feedback