·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
SkillOpt: Executive Strategy for Self-Evolving Agent Skills3h◆When Does Synthetic Patent Data Help? Volume-Fidelity Trade-offs in Low-Resource Multi-Label Classification3h◆Advancing Graph Few-Shot Learning via In-Context Learning3h◆The Model Is Not the Product: A Dual-Pillar Architecture for Local-First Psychological Coaching3h◆JT-SAFE-V2: Safety-by-Design Foundation Model with World-Context Data3h◆AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning3h◆Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs3h◆Hypothesis Generation and Inductive Inference in Children and Language Models3h◆DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations3h◆Jailbreak to Protect: Buffering and Reinforcing via Temporary Jailbreaking for Safe Fine-Tuning in Large Language Models3h◆VectorArk: Learning Practical Image Vectorization with Rounded Polygon Representation3h◆Summoning the Oracle to Slay It: Mitigating Look-Ahead Bias in Financial Backtesting with Large Language Models3h◆Hera: Learning Long-Horizon Coordination for Device-Cloud Collaborative LLM Agents3h◆GlobalDentBench: A Multinational Benchmark for Evaluating LLM Clinical Reasoning in Dentistry with Expert Calibration3h◆Beyond Inference-Only Deployment: Comparing Weight-Based Consolidation Against Cascading Compaction3h◆MDIA: A Multi-Agent Diagnostic Intelligence Pipeline on HealthBench Professional3h◆Fundamental Limitation in Explaining AI3h◆Balancing Fairness, Privacy, and Accuracy: A Multitask Adversarial Framework for Centralized Data-Driven Systems3h◆Hylos: Operability Contracts for Model-Native Spatial Intelligence3h◆CoRe-Code: Collaborative Reinforcement Learning for Code Generation3h◆SkillOpt: Executive Strategy for Self-Evolving Agent Skills3h◆When Does Synthetic Patent Data Help? Volume-Fidelity Trade-offs in Low-Resource Multi-Label Classification3h◆Advancing Graph Few-Shot Learning via In-Context Learning3h◆The Model Is Not the Product: A Dual-Pillar Architecture for Local-First Psychological Coaching3h◆JT-SAFE-V2: Safety-by-Design Foundation Model with World-Context Data3h◆AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning3h◆Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs3h◆Hypothesis Generation and Inductive Inference in Children and Language Models3h◆DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations3h◆Jailbreak to Protect: Buffering and Reinforcing via Temporary Jailbreaking for Safe Fine-Tuning in Large Language Models3h◆VectorArk: Learning Practical Image Vectorization with Rounded Polygon Representation3h◆Summoning the Oracle to Slay It: Mitigating Look-Ahead Bias in Financial Backtesting with Large Language Models3h◆Hera: Learning Long-Horizon Coordination for Device-Cloud Collaborative LLM Agents3h◆GlobalDentBench: A Multinational Benchmark for Evaluating LLM Clinical Reasoning in Dentistry with Expert Calibration3h◆Beyond Inference-Only Deployment: Comparing Weight-Based Consolidation Against Cascading Compaction3h◆MDIA: A Multi-Agent Diagnostic Intelligence Pipeline on HealthBench Professional3h◆Fundamental Limitation in Explaining AI3h◆Balancing Fairness, Privacy, and Accuracy: A Multitask Adversarial Framework for Centralized Data-Driven Systems3h◆Hylos: Operability Contracts for Model-Native Spatial Intelligence3h◆CoRe-Code: Collaborative Reinforcement Learning for Code Generation3h◆
News/On the Impact of Class Imbalance on the Learning Dynamics of Deep Neural Networks:An Intuitive Insight
arxiv
PublishedMay 26, 2026 at 4:00 AM

On the Impact of Class Imbalance on the Learning Dynamics of Deep Neural Networks:An Intuitive Insight

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

arXiv:2605.24908v1 Announce Type: cross Abstract: Class imbalance in deep neural networks (DNNs) has witnessed a rapid increase in research attention in recent years. However, the varying accounts of the reasons behind the poor performance of DNN on imbalance data in pertinent literature shows that

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
arxivSkillOpt: Executive Strategy for Self-Evolving Agent Skills3harxivWhen Does Synthetic Patent Data Help? Volume-Fidelity Trade-offs in Low-Resource Multi-Label Classification3harxivAdvancing Graph Few-Shot Learning via In-Context Learning3harxivThe Model Is Not the Product: A Dual-Pillar Architecture for Local-First Psychological Coaching3h
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