·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Same Task, Different Statement: When and How Reformulation Helps LLM Problem Solving5h◆EvoX: Rethinking the Agent Harness as a Runtime Policy5h◆Benchmarking AI risk behaviors5h◆Singularity-Bench: Can Agents Build Agents?5h◆Folio: Recoverable Multi-Resolution Context Folding for Long-Horizon Language Agents5h◆Safety as a Budget: Calibrating the Control Channel of Constrained GRPO for Vision-Language Models5h◆Cortex-FL: Agent-Mediated Collaboration for Federated Learning5h◆Al-Ghasali: A Computational Framework for Simulating Islamic Jurisprudential Rulings via Finite State Automata and Large Language Models5h◆Ballad: Bandit-Based LLM Routing for Automated Heuristic Discovery5h◆When Safety Becomes an Attack Surface: Manipulating Abstention in Tool-Using AI Agents5h◆LLM-Enhanced Deep Reinforcement Learning with Human-in-the-Loop Calibration for Real-Time Optimal Power Flow5h◆A Cardiorespiratory Foundation Model Pretrained on Ten Physiological Signals for Perioperative and Critical Care5h◆Beyond Named Mechanisms: Economic Institutions as a Design Space for Artificial Agents5h◆SkinAdvisor: An AI-Driven Decision Framework for Skin Production Planning via Multi-Modal Transformer Sales Forecasting5h◆Estimating Image Creativity5h◆Quantum Recurrent Architectures for Financial Time Series Prediction5h◆Matrix-Based Ant Colony Search with Batched Decoding for Hierarchical Replenishment Routing: A Stylized Case Study Based on Korean City Data5h◆Breath-Dx: A Time Series Classification Benchmark for Disease Screening with eNose5h◆Where Is Uncertainty Observed? A Mechanistic Understanding of Uncertainty Quantification in LLMs5h◆Latent Imagination for Parameterized Action Spaces5h◆Same Task, Different Statement: When and How Reformulation Helps LLM Problem Solving5h◆EvoX: Rethinking the Agent Harness as a Runtime Policy5h◆Benchmarking AI risk behaviors5h◆Singularity-Bench: Can Agents Build Agents?5h◆Folio: Recoverable Multi-Resolution Context Folding for Long-Horizon Language Agents5h◆Safety as a Budget: Calibrating the Control Channel of Constrained GRPO for Vision-Language Models5h◆Cortex-FL: Agent-Mediated Collaboration for Federated Learning5h◆Al-Ghasali: A Computational Framework for Simulating Islamic Jurisprudential Rulings via Finite State Automata and Large Language Models5h◆Ballad: Bandit-Based LLM Routing for Automated Heuristic Discovery5h◆When Safety Becomes an Attack Surface: Manipulating Abstention in Tool-Using AI Agents5h◆LLM-Enhanced Deep Reinforcement Learning with Human-in-the-Loop Calibration for Real-Time Optimal Power Flow5h◆A Cardiorespiratory Foundation Model Pretrained on Ten Physiological Signals for Perioperative and Critical Care5h◆Beyond Named Mechanisms: Economic Institutions as a Design Space for Artificial Agents5h◆SkinAdvisor: An AI-Driven Decision Framework for Skin Production Planning via Multi-Modal Transformer Sales Forecasting5h◆Estimating Image Creativity5h◆Quantum Recurrent Architectures for Financial Time Series Prediction5h◆Matrix-Based Ant Colony Search with Batched Decoding for Hierarchical Replenishment Routing: A Stylized Case Study Based on Korean City Data5h◆Breath-Dx: A Time Series Classification Benchmark for Disease Screening with eNose5h◆Where Is Uncertainty Observed? A Mechanistic Understanding of Uncertainty Quantification in LLMs5h◆Latent Imagination for Parameterized Action Spaces5h◆
News/When Is Proof-Step Verification Worth Its Cost? Operating Requirements of Certify-or-Decline Verification for Open-Weight Mathematical Reasoning
arxiv
PublishedNovember 14, 2026 at 5:00 AM

When Is Proof-Step Verification Worth Its Cost? Operating Requirements of Certify-or-Decline Verification for Open-Weight Mathematical Reasoning

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

arXiv:2611.14794v1 Announce Type: new Abstract: Proof-step verification rewrites a solver's answer into typed steps and uses LLM judges to certify or decline. We evaluate this pipeline on open-weight models across GPQA, AIME-2025, and a MATH-500 subset. Certification raises answer-key-match precisio

Stay posted· Newsletter

Three short emails a week: Monday, Wednesday and Friday. Top movers, new models and the most-referenced story.

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

Your email is used only to send The Bubble Brief. Privacy policy

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
arxivSame Task, Different Statement: When and How Reformulation Helps LLM Problem Solving5harxivEvoX: Rethinking the Agent Harness as a Runtime Policy5harxivBenchmarking AI risk behaviors5harxivSingularity-Bench: Can Agents Build Agents?5h
The Bubble Brief
MON · WED · FRI

Read AI insights every Monday, Wednesday and Friday — top movers, new models, story of the week.

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

Your email is used only to send The Bubble Brief. Privacy policy

Originally published on arxiv ↗
PrivacyTermsRefundsLegal notice|AboutMethodologyContact

© 2026 Databubble

Built by Marouane Gazouzi · Readyo, AI interview coach

HomeModelsNews