·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Effects of sparsity and superposition on loss in simple autoencoders3h◆Towards Anomaly Detection on Relational Data3h◆PACT: Preserving Anchored Cores in Task-vectors for Model Merging3h◆MetaboNet-Bench: A Multi-modal Benchmark for Glucose Forecasting in Type 1 Diabetes3h◆BLADE: Scalable Bi-level Adaptive Data Selection for LLM Training3h◆InTrain: Intrinsic Trainability for Zero-Cost Neural Architecture Search3h◆Robust and Interpretable Adaptation of Equivariant Materials Foundation Models via Sparsity-promoting Fine-tuning3h◆Contextualizing Biological Language Models across Modalities via Logit-Space Contrastive Alignment3h◆Trainable Photonic Measurement for Physics-Informed PDE Learning3h◆Low-Cost Neuromorphic Fall Detection Using Synthetic Event Data and Hybrid SNNs3h◆RouteJudge: An Open Platform for Reproducible and Preference-Aware LLM Routing3h◆Online Distributional Prediction via Latent Cluster Geometry Under Drift and Corruption3h◆Seed-Guided Semi-Supervised Clustering by A-Contrario Anomaly Detection3h◆Identifying Structural Biases from Causal Mechanism Shifts3h◆Semantic Robustness Certification for Vision-Language Models3h◆Smoothness-Based Derandomization of PAC-Bayes Bounds3h◆JourneyFormer: Encoding Airbnb Guest Journey with Sequence Modeling3h◆Seeing Before Reasoning: Decoupling Perception and Reasoning for Shortcut-Resilient Multimodal On-Policy Self-Distillation3h◆INDEQS: Informed Neural controlled Differential EQuationS3h◆ChronoSurv: A Clinical Pathway-Guided Graph Framework for Multimodal Survival Analysis3h◆Effects of sparsity and superposition on loss in simple autoencoders3h◆Towards Anomaly Detection on Relational Data3h◆PACT: Preserving Anchored Cores in Task-vectors for Model Merging3h◆MetaboNet-Bench: A Multi-modal Benchmark for Glucose Forecasting in Type 1 Diabetes3h◆BLADE: Scalable Bi-level Adaptive Data Selection for LLM Training3h◆InTrain: Intrinsic Trainability for Zero-Cost Neural Architecture Search3h◆Robust and Interpretable Adaptation of Equivariant Materials Foundation Models via Sparsity-promoting Fine-tuning3h◆Contextualizing Biological Language Models across Modalities via Logit-Space Contrastive Alignment3h◆Trainable Photonic Measurement for Physics-Informed PDE Learning3h◆Low-Cost Neuromorphic Fall Detection Using Synthetic Event Data and Hybrid SNNs3h◆RouteJudge: An Open Platform for Reproducible and Preference-Aware LLM Routing3h◆Online Distributional Prediction via Latent Cluster Geometry Under Drift and Corruption3h◆Seed-Guided Semi-Supervised Clustering by A-Contrario Anomaly Detection3h◆Identifying Structural Biases from Causal Mechanism Shifts3h◆Semantic Robustness Certification for Vision-Language Models3h◆Smoothness-Based Derandomization of PAC-Bayes Bounds3h◆JourneyFormer: Encoding Airbnb Guest Journey with Sequence Modeling3h◆Seeing Before Reasoning: Decoupling Perception and Reasoning for Shortcut-Resilient Multimodal On-Policy Self-Distillation3h◆INDEQS: Informed Neural controlled Differential EQuationS3h◆ChronoSurv: A Clinical Pathway-Guided Graph Framework for Multimodal Survival Analysis3h◆
News/Noise-Driven Escape from Metastable Phases explains Grokking in Deep Neural Networks
arxiv
PublishedJune 17, 2026 at 4:00 AM
—neutral

Noise-Driven Escape from Metastable Phases explains Grokking in Deep Neural Networks

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

arXiv:2606.17120v1 Announce Type: new Abstract: Deep neural networks (DNNs) exhibit first order phase transitions under variations of the L2 regularization strength, with each transition marking the onset of a new learnable feature. Below a critical regularization strength, all features are in princ

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
arxivEffects of sparsity and superposition on loss in simple autoencoders3harxivTowards Anomaly Detection on Relational Data3harxivPACT: Preserving Anchored Cores in Task-vectors for Model Merging3harxivMetaboNet-Bench: A Multi-modal Benchmark for Glucose Forecasting in Type 1 Diabetes3h
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