·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Perplexity trusts GPT-6 Astra with end-to-end systems-2455m◆TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents3h◆Natural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query Generation3h◆SPECTRA: Band-Routed Embedding and Stage-Wise LoRA for Cross-Sensor Fine-Tuning of Geospatial Foundation Models3h◆Towards a Deterministic Math Solver for Clinical Language Models3h◆Demystifying the Privacy-Utility Trade-off in LLM Interactions3h◆Autonomous Chemical Mechanistic Discovery through Agentic Reasoning and Validation3h◆DRG-MAPPO: Hierarchical Dynamic Role-Graph Multi-Agent Reinforcement Learning for Cooperative Air Combat3h◆NovGauge: A Fine-Grained Benchmark for Diagnosing LLMs' Capability in Paper Novelty Assessment3h◆RAMamba-Net: A Reliability-Aware and Mamba-Based Multimodal Fusion Network for Auditory Attention Detection3h◆Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning3h◆From Queries to Narratives: Cultural Heritage Data Stories for Knowledge Graph Exploration and Quality Assessment3h◆LLMs as Post-hoc Auditors of Physiological Plausibility in Symbolic Regression: A Clinician-Evaluated Case Study3h◆Calibration-Aware Uncertainty Cascades for Efficient Heterogeneous Model Collaboration3h◆RouteRepair: Instance-Level Failure Diagnosis and Targeted Repair in LLM-Based Automated Heuristic Design for Routing Optimization3h◆Flexible and Interpretable Accent Distance Measurements3h◆The Convention Gap: Towards Measuring Implicit Communication in Cooperative AI Evaluation3h◆From Document Silos to Process Intelligence: A Multi-Layer Knowledge Graph for CMC Process Development3h◆ActMap: Single-Pass Uncertainty Quantification from Generation-Time Activation Maps3h◆Extending SMT Solving with Non-Ground Clause Learning3h◆Perplexity trusts GPT-6 Astra with end-to-end systems-2455m◆TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents3h◆Natural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query Generation3h◆SPECTRA: Band-Routed Embedding and Stage-Wise LoRA for Cross-Sensor Fine-Tuning of Geospatial Foundation Models3h◆Towards a Deterministic Math Solver for Clinical Language Models3h◆Demystifying the Privacy-Utility Trade-off in LLM Interactions3h◆Autonomous Chemical Mechanistic Discovery through Agentic Reasoning and Validation3h◆DRG-MAPPO: Hierarchical Dynamic Role-Graph Multi-Agent Reinforcement Learning for Cooperative Air Combat3h◆NovGauge: A Fine-Grained Benchmark for Diagnosing LLMs' Capability in Paper Novelty Assessment3h◆RAMamba-Net: A Reliability-Aware and Mamba-Based Multimodal Fusion Network for Auditory Attention Detection3h◆Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning3h◆From Queries to Narratives: Cultural Heritage Data Stories for Knowledge Graph Exploration and Quality Assessment3h◆LLMs as Post-hoc Auditors of Physiological Plausibility in Symbolic Regression: A Clinician-Evaluated Case Study3h◆Calibration-Aware Uncertainty Cascades for Efficient Heterogeneous Model Collaboration3h◆RouteRepair: Instance-Level Failure Diagnosis and Targeted Repair in LLM-Based Automated Heuristic Design for Routing Optimization3h◆Flexible and Interpretable Accent Distance Measurements3h◆The Convention Gap: Towards Measuring Implicit Communication in Cooperative AI Evaluation3h◆From Document Silos to Process Intelligence: A Multi-Layer Knowledge Graph for CMC Process Development3h◆ActMap: Single-Pass Uncertainty Quantification from Generation-Time Activation Maps3h◆Extending SMT Solving with Non-Ground Clause Learning3h◆
News/Output Embedding Centering for Stable LLM Pretraining
arxiv
PublishedSeptember 12, 2026 at 4:00 AM
—neutral

Output Embedding Centering for Stable LLM Pretraining

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

arXiv:2601.02031v3 Announce Type: replace-cross Abstract: Pretraining of large language models is not only expensive but also prone to certain training instabilities. A specific instability that often occurs at the end of training is output logit divergence. The most widely used mitigation strategie

Stay posted· Newsletter

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

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

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
arxivTeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents3harxivNatural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query Generation3harxivSPECTRA: Band-Routed Embedding and Stage-Wise LoRA for Cross-Sensor Fine-Tuning of Geospatial Foundation Models3harxivTowards a Deterministic Math Solver for Clinical Language Models3h
The Bubble Brief
WEEKLY

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

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

Originally published on arxiv ↗
HomeModelsNews