·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Connected by Construction: Learning Tractable Near-Tour Marginals for Traveling Salesman Problems3h◆Good Benchmarks3h◆On-Device Deep Research at 4B: Exposure Bounds Faithfulness, Retrieval Bounds Coverage3h◆How Many Tasks Are Enough for Agent Benchmark Decisions? A Replay Analysis of Public LLM Agent Benchmarks3h◆PM-Bench: Evaluating Prospective Memory in LLM Agents3h◆Isolation as a First-Class Principle for LLM-Agent System Safety: Concepts, Taxonomy, Challenges and Future Directions3h◆Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models3h◆FormalAnalyticGeo: A Neural-Symbolic Based Framework for Multimodal Analytic Geometry Problem Generation3h◆Burst Spiking Neural Networks3h◆Gene Expression-Informed Jointly Controlled Generative Modeling for Precision Molecular Design3h◆Removable Defects: The Economics and Limits of Deliberate Deficiency3h◆Sparse Inter-Layer Dependencies of Transformer FFN Neurons3h◆Are we Merging the Right Models? Impact of Expert Training Duration on Model Merging for LLMs3h◆HPC-Enabled Video-based Coastal Wave Parameter Estimation Using V-JEPA and Deep Spatiotemporal Learning3h◆Sparse Autoencoders for Interpretable Out-of-Distribution Detection3h◆OOD-RL-Bench: A Benchmark Framework for Out-of-Distribution Detection in Reinforcement Learning3h◆Bulkhead: Automated Semantic Detection and Remediation of Container Escape Vulnerabilities3h◆Learning-based Probabilistic Load Forecasting with Post-hoc and In-model Uncertainty3h◆Accelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques3h◆UR-VC: Unsupervised Robotic Value Correction for Time-Derived Progress Proxies3h◆Connected by Construction: Learning Tractable Near-Tour Marginals for Traveling Salesman Problems3h◆Good Benchmarks3h◆On-Device Deep Research at 4B: Exposure Bounds Faithfulness, Retrieval Bounds Coverage3h◆How Many Tasks Are Enough for Agent Benchmark Decisions? A Replay Analysis of Public LLM Agent Benchmarks3h◆PM-Bench: Evaluating Prospective Memory in LLM Agents3h◆Isolation as a First-Class Principle for LLM-Agent System Safety: Concepts, Taxonomy, Challenges and Future Directions3h◆Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models3h◆FormalAnalyticGeo: A Neural-Symbolic Based Framework for Multimodal Analytic Geometry Problem Generation3h◆Burst Spiking Neural Networks3h◆Gene Expression-Informed Jointly Controlled Generative Modeling for Precision Molecular Design3h◆Removable Defects: The Economics and Limits of Deliberate Deficiency3h◆Sparse Inter-Layer Dependencies of Transformer FFN Neurons3h◆Are we Merging the Right Models? Impact of Expert Training Duration on Model Merging for LLMs3h◆HPC-Enabled Video-based Coastal Wave Parameter Estimation Using V-JEPA and Deep Spatiotemporal Learning3h◆Sparse Autoencoders for Interpretable Out-of-Distribution Detection3h◆OOD-RL-Bench: A Benchmark Framework for Out-of-Distribution Detection in Reinforcement Learning3h◆Bulkhead: Automated Semantic Detection and Remediation of Container Escape Vulnerabilities3h◆Learning-based Probabilistic Load Forecasting with Post-hoc and In-model Uncertainty3h◆Accelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques3h◆UR-VC: Unsupervised Robotic Value Correction for Time-Derived Progress Proxies3h◆
News/CAST: Non-Privileged Clipped Asymmetric Self-Teaching with Advantage Flipping for GRPO
arxiv
PublishedJune 2, 2026 at 4:00 AM

CAST: Non-Privileged Clipped Asymmetric Self-Teaching with Advantage Flipping for GRPO

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

arXiv:2606.00172v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR), especially Group Relative Policy Optimization (GRPO), has been widely used to improve reasoning in large language models. However, outcome-level rewards provide only sparse supervision, and group-r

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
arxivConnected by Construction: Learning Tractable Near-Tour Marginals for Traveling Salesman Problems3harxivGood Benchmarks3harxivOn-Device Deep Research at 4B: Exposure Bounds Faithfulness, Retrieval Bounds Coverage3harxivHow Many Tasks Are Enough for Agent Benchmark Decisions? A Replay Analysis of Public LLM Agent Benchmarks3h
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