·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
VibeJam: A User Study Platform for Web Development with Agents4h◆Small Language Models as Judges for Rubric-Based Reinforcement Learning4h◆How Order-Sensitive Are LLMs? OrderProbe for Deterministic Structural Reconstruction4h◆CoReflect: A Reflective Co-Evolution Framework for Improving Conversational Evaluation4h◆EMemBench: Interactive Benchmarking of Episodic Memory for VLM Agents4h◆Constrained Group Relative Policy Optimization4h◆Propensity Straight-Through Gradients for Discrete Stochastic Systems4h◆GFlowNets and variational inference4h◆Loci Similes: A Benchmark for Extracting Intertextualities in Latin Literature4h◆PRACTICE: From Experience to Expertise in Self-Evolving Embodied Agents4h◆Compact and Infinite-Order Error Analysis for Null-Space SVD Estimation4h◆Large language model-enabled automated data extraction for concrete materials informatics4h◆PlanCraft: Sketch, Refine, and Furnish for Architect-Inspired Progressive 3D Residential Scene Generation4h◆Learning to Look Again: Loss-Gap Supervision for Free-form Crop Routing in Vision-Language Models4h◆BCPPO: Bachelier-Inspired Constrained Proximal Policy Optimization for Tail-Risk-Aware Safe Reinforcement Learning4h◆Enhancing Low-Resource Language Reasoning via High-Resource Language Feature Transfer4h◆PropUQ-MAS: Propagation-Aware Uncertainty Quantification for LLM Multi-Agent Systems4h◆Tensor Methods for Language Models: From Token Representation to Training, Adaptation, Inference, Compression, and Interpretability4h◆Lot Machine: Multimodal Lot Extraction from Auction Catalogs4h◆Trajectory-Initialized Neural Double Q-Routing for Large-Scale Overhead Hoist Transport Systems4h◆VibeJam: A User Study Platform for Web Development with Agents4h◆Small Language Models as Judges for Rubric-Based Reinforcement Learning4h◆How Order-Sensitive Are LLMs? OrderProbe for Deterministic Structural Reconstruction4h◆CoReflect: A Reflective Co-Evolution Framework for Improving Conversational Evaluation4h◆EMemBench: Interactive Benchmarking of Episodic Memory for VLM Agents4h◆Constrained Group Relative Policy Optimization4h◆Propensity Straight-Through Gradients for Discrete Stochastic Systems4h◆GFlowNets and variational inference4h◆Loci Similes: A Benchmark for Extracting Intertextualities in Latin Literature4h◆PRACTICE: From Experience to Expertise in Self-Evolving Embodied Agents4h◆Compact and Infinite-Order Error Analysis for Null-Space SVD Estimation4h◆Large language model-enabled automated data extraction for concrete materials informatics4h◆PlanCraft: Sketch, Refine, and Furnish for Architect-Inspired Progressive 3D Residential Scene Generation4h◆Learning to Look Again: Loss-Gap Supervision for Free-form Crop Routing in Vision-Language Models4h◆BCPPO: Bachelier-Inspired Constrained Proximal Policy Optimization for Tail-Risk-Aware Safe Reinforcement Learning4h◆Enhancing Low-Resource Language Reasoning via High-Resource Language Feature Transfer4h◆PropUQ-MAS: Propagation-Aware Uncertainty Quantification for LLM Multi-Agent Systems4h◆Tensor Methods for Language Models: From Token Representation to Training, Adaptation, Inference, Compression, and Interpretability4h◆Lot Machine: Multimodal Lot Extraction from Auction Catalogs4h◆Trajectory-Initialized Neural Double Q-Routing for Large-Scale Overhead Hoist Transport Systems4h◆
News/From SGD to Muon: Adaptive Optimization via Schatten-p Norms
arxiv
PublishedMay 21, 2026 at 4:00 AM
▲bullish

From SGD to Muon: Adaptive Optimization via Schatten-p Norms

Source
arxiv.orgfull article ↗
Read on arxiv→
Publisher summary· verbatim

arXiv:2605.19781v1 Announce Type: new Abstract: Modern optimizers, like Muon, impose matrix-wise geometry constraints on their updates. These matrix-wise constraints can be unified under Linear Minimization Oracle (LMO) theory. However, all current methods impose fixed LMO geometries for the update

Stay posted· Newsletter

A 5-min weekly brief — top movers, price watch, story of the week.

// no spam · unsubscribe one-click · free forever

Discussion
Mentioned models
05
  • 01
    Muon
  • 02
    SGD
  • 03
    Adam
  • 04
    AdamW
  • 05
    MuAdam
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
04
#optimization#deep learning#neural networks#artificial intelligence

No replies yet. Be first.

Mentioned models
05
  • 01
    Muon
  • 02
    SGD
  • 03
    Adam
  • 04
    AdamW
  • 05
    MuAdam
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
04
#optimization#deep learning#neural networks#artificial intelligence

Related coverage

More from ARXIV
arxivVibeJam: A User Study Platform for Web Development with Agents4harxivSmall Language Models as Judges for Rubric-Based Reinforcement Learning4harxivHow Order-Sensitive Are LLMs? OrderProbe for Deterministic Structural Reconstruction4harxivCoReflect: A Reflective Co-Evolution Framework for Improving Conversational Evaluation4h
The Bubble Brief
WEEKLY

Read optimization insights every Tuesday — top movers, new releases, story of the week.

// no spam · unsubscribe one-click · free forever

Originally published on arxiv ↗
HomeModelsNews