·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster38m◆AI promises productivity, but for managers, it’s mostly a headache for now1h◆OpenAI says an AI agent lied about disabling its safety monitoring system1h◆ChatGPT Ads expands to 26 new countries as over 100,000 advertisers have used the platform2h◆NVIDIA Kumo Forecast Combines Pretrained Time-Series Representations with Historical Pattern Retrieval and Interpretability2h◆OpenAI models are coming to Google Cloud3h◆Announcing SimReady Central: An Open Library of Simulation-Ready 3D Assets for Robotics and AI Factories4h◆How we built ChatGPT’s novel ad stack for user trust and relevance8h◆Using thousands of AI agents to uncover new leads in biological data10h◆An AI Security Agent for University ACMIS: Multi-Vector Threat Detection and Automated Response11h◆Learning Transferable Latent User Preferences for Human-Aligned Decision Making11h◆Repair the Core, Not the Symptom: Planning-Graph Repair for Long-Horizon Agents11h◆IndustryForge-27B: A Domain-Enhanced Multimodal Foundation Model for Industrial CAD11h◆Ethical LLM-Assisted Research: A Framework for Responsible Delegation, Verification, and Epistemic Value11h◆PALATE: Peculiar Application of the Law of Total Expectation to Enhance the Evaluation of Deep Generative Models11h◆WoW-Bench: Evaluating Fine-Grained Acoustic Perception in Audio-Language Models via Marine Mammal Vocalizations11h◆When Synthetic Speech Is All You Have: Better Call GRPO11h◆Reducing belief in conspiracy theories as they unfold using large language models11h◆Harnessing Magnitude-Only and Complex Measurements for Improved Dynamic MRI Reconstruction with Learned Priors11h◆EDISCO: Equivariant DIScrete Diffusion for Euclidean Combinatorial Optimization11h◆Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster38m◆AI promises productivity, but for managers, it’s mostly a headache for now1h◆OpenAI says an AI agent lied about disabling its safety monitoring system1h◆ChatGPT Ads expands to 26 new countries as over 100,000 advertisers have used the platform2h◆NVIDIA Kumo Forecast Combines Pretrained Time-Series Representations with Historical Pattern Retrieval and Interpretability2h◆OpenAI models are coming to Google Cloud3h◆Announcing SimReady Central: An Open Library of Simulation-Ready 3D Assets for Robotics and AI Factories4h◆How we built ChatGPT’s novel ad stack for user trust and relevance8h◆Using thousands of AI agents to uncover new leads in biological data10h◆An AI Security Agent for University ACMIS: Multi-Vector Threat Detection and Automated Response11h◆Learning Transferable Latent User Preferences for Human-Aligned Decision Making11h◆Repair the Core, Not the Symptom: Planning-Graph Repair for Long-Horizon Agents11h◆IndustryForge-27B: A Domain-Enhanced Multimodal Foundation Model for Industrial CAD11h◆Ethical LLM-Assisted Research: A Framework for Responsible Delegation, Verification, and Epistemic Value11h◆PALATE: Peculiar Application of the Law of Total Expectation to Enhance the Evaluation of Deep Generative Models11h◆WoW-Bench: Evaluating Fine-Grained Acoustic Perception in Audio-Language Models via Marine Mammal Vocalizations11h◆When Synthetic Speech Is All You Have: Better Call GRPO11h◆Reducing belief in conspiracy theories as they unfold using large language models11h◆Harnessing Magnitude-Only and Complex Measurements for Improved Dynamic MRI Reconstruction with Learned Priors11h◆EDISCO: Equivariant DIScrete Diffusion for Euclidean Combinatorial Optimization11h◆
News/Dense vs Sparse Pretraining at Tiny Scale: Active-Parameter vs Total-Parameter Matching
arxiv
PublishedMay 14, 2026 at 4:00 AM

Dense vs Sparse Pretraining at Tiny Scale: Active-Parameter vs Total-Parameter Matching

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

arXiv:2605.13769v1 Announce Type: cross Abstract: We study dense and mixture-of-experts (MoE) transformers in a tiny-scale pretraining regime under a shared LLaMA-style decoder training recipe. The sparse model replaces dense feed-forward blocks with Mixtral-style routed experts. Dense baselines are

Stay posted· Newsletter

Three short emails a week: Monday, Wednesday and Friday. Top movers, new models and the most-referenced story.

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

Your email is used only to send The Bubble Brief. Privacy policy

Discussion
Source
↗
arxiv
Read original ↗All from arxiv →

No replies yet. Be first.

Source
↗
arxiv
Read original ↗All from arxiv →

Related coverage

More from ARXIV
arxivAn AI Security Agent for University ACMIS: Multi-Vector Threat Detection and Automated Response11harxivLearning Transferable Latent User Preferences for Human-Aligned Decision Making11harxivRepair the Core, Not the Symptom: Planning-Graph Repair for Long-Horizon Agents11harxivIndustryForge-27B: A Domain-Enhanced Multimodal Foundation Model for Industrial CAD11h
The Bubble Brief
MON · WED · FRI

Read AI insights every Monday, Wednesday and Friday — top movers, new models, story of the week.

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

Your email is used only to send The Bubble Brief. Privacy policy

Originally published on arxiv ↗
PrivacyTermsRefundsLegal notice|AboutMethodologyContact

© 2026 Databubble

Built by Marouane Gazouzi · Readyo, AI interview coach

HomeModelsNews