·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
OpenAI’s new agent is a shot at Meta — but can it compete with free?1h◆Photon held a funeral for mobile apps. Now it has $4.5M to help replace them with agents.1h◆Inside our months-long investigation into Kevin O’Leary’s Utah data center debacle1h◆Hearing tech startup Legato launches its AI hearing glasses2h◆Satlyt, founded by a former Google and SpaceX product manager, raises $8M to run AI on satellites3h◆Reasoning Externalization for Faithful Large Language Model Narratives of Stock Return Predictions11h◆Consistent Plan-Act for Long-Horizon Agentic Tasks11h◆Predictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance11h◆Boosting Adversarial Robustness and Generalization with Dictionary Structure11h◆PivotOPD: Learning to Recover from Pivotal Mistakes in Multi-Turn Agents11h◆TED:Text-Axis Evidence Decomposition for Prompted Anomaly Localization11h◆COMiT: Learning Structured Visual Tokens through Sequential Communication11h◆On the (In)effectiveness of AMR Augmentation for Large Language Models11h◆cua-speedrun: Standardized Benchmarking of the Speed of Computer-Use Agents11h◆Compress to Remember: Learning Compact Memory via On-Policy Distillation for Long Video Generation11h◆Building Socio-Affective Artificial Intelligence for Interactive Multi-Agent Simulations11h◆Mitigating Memorization In Language Models11h◆MoEless: Efficient MoE LLM Serving with Serverless Experts11h◆Fork-Think with Confidence11h◆Where Does Randomness Matter in Neural Cellular Automata?11h◆OpenAI’s new agent is a shot at Meta — but can it compete with free?1h◆Photon held a funeral for mobile apps. Now it has $4.5M to help replace them with agents.1h◆Inside our months-long investigation into Kevin O’Leary’s Utah data center debacle1h◆Hearing tech startup Legato launches its AI hearing glasses2h◆Satlyt, founded by a former Google and SpaceX product manager, raises $8M to run AI on satellites3h◆Reasoning Externalization for Faithful Large Language Model Narratives of Stock Return Predictions11h◆Consistent Plan-Act for Long-Horizon Agentic Tasks11h◆Predictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance11h◆Boosting Adversarial Robustness and Generalization with Dictionary Structure11h◆PivotOPD: Learning to Recover from Pivotal Mistakes in Multi-Turn Agents11h◆TED:Text-Axis Evidence Decomposition for Prompted Anomaly Localization11h◆COMiT: Learning Structured Visual Tokens through Sequential Communication11h◆On the (In)effectiveness of AMR Augmentation for Large Language Models11h◆cua-speedrun: Standardized Benchmarking of the Speed of Computer-Use Agents11h◆Compress to Remember: Learning Compact Memory via On-Policy Distillation for Long Video Generation11h◆Building Socio-Affective Artificial Intelligence for Interactive Multi-Agent Simulations11h◆Mitigating Memorization In Language Models11h◆MoEless: Efficient MoE LLM Serving with Serverless Experts11h◆Fork-Think with Confidence11h◆Where Does Randomness Matter in Neural Cellular Automata?11h◆
News/Selective Off-Policy Reference Tuning with Plan Guidance
arxiv
PublishedSeptember 29, 2026 at 4:00 AM
—neutral

Selective Off-Policy Reference Tuning with Plan Guidance

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

arXiv:2605.11505v3 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards helps reasoning, but GRPO-style methods stall on hard prompts where all sampled rollouts fail. SORT adds a repair update for those failures without changing rollout generation: it derives a plan from t

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
arxivReasoning Externalization for Faithful Large Language Model Narratives of Stock Return Predictions11harxivConsistent Plan-Act for Long-Horizon Agentic Tasks11harxivPredictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance11harxivBoosting Adversarial Robustness and Generalization with Dictionary Structure11h
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 ↗
Built by Marouane Gazouzi
HomeModelsNews