arXiv AI August 12, 2026 By Cheng Yan, Zhijun Fan, Guangyang Ye, Fan Xu, Xiang Xia, Yawei Wang, Wuyang Zhang

UPAIR: Diagnosing Reasoning States via Uncertainty-Progress Alignment for Selective Intervention

arXiv:2607. 17188v2 Announce Type: replace Abstract: While test-time scaling improves the problem-solving ability of large reasoning models (LRMs) through additional inference-time computation, it can also exacerbate overthinking and underthinking, which we formulate as reasoning state--action mismatch.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

arXiv AI

Jul 21

Is Your Model Thinking or Just Stagnating? PUMA: Diagnosing Reasoning Pathology via Phase-Momentum Alignment

arXiv:2607. 17188v1 Announce Type: new Abstract: Test-time scaling empowers Large Reasoning Models (LRMs) to tackle complex tasks via extensive Chain-of-Thought (CoT).

By Cheng Yan, Guangyang Ye, Wuyang Zhang, Fan Xu, Zhijun Fan, Xiang Xia, Yanyong Zhang

benchmarks safety

arXiv AI

1d ago

CEDAR-GRPO: Process-Aware Reinforcement Learning for General Abductive Reasoning in LLMs

arXiv:2608. 14791v1 Announce Type: new Abstract: Abductive reasoning, often characterized as inference to the best explanation, is central to explanation under uncertainty, from everyday sense-making and investigation to scientific discovery.

By Moein Salimi, Danial Parnian, Shaygan Adim, Amirmohammad Ebrahiminasab, Nima Alighardashi, Parsa Gholami, Sahand Akramipour, Mahdi Jafari Siavoshani, Mohammad Hossein Rohban

llms reinforcement-learning benchmarks

arXiv AI

1d ago

A Temporal Reasoning Benchmarking Framework for LRMs via Difficulty-controlled and Dynamic Test Generation

arXiv:2607. 04784v1 Announce Type: cross Abstract: Defining the reasoning boundaries and ensuring the reliability of Large Reasoning Models (LRMs) remains a critical challenge.

By Shide Zhou, Kailong Wang, Ling Shi, Haoyu Wang

benchmarks

arXiv AI

1d ago

Divergent-Convergent Reasoning: Scaling Test-Time Compute through Structured Solution Synthesis

arXiv:2608. 15303v1 Announce Type: new Abstract: Test-time compute can substantially improve Large Language Model (LLM) reasoning performance, yet how and when additional compute helps remains poorly understood.

By Bo Wen, Yuhao Chen, Erhan Bilal, Carla Agurto Rios, Chen Wang, Junchen Jiang

llms agents

arXiv AI

Jun 2

Process Reward Agents for Steering Knowledge-Intensive Reasoning

arXiv:2604. 09482v2 Announce Type: replace Abstract: Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources.

By Jiwoong Sohn, Tomasz Sternal, Kenneth Styppa, Torsten Hoefler, Michael Moor

rag agents benchmarks

arXiv AI

Jul 21

Can We Break LLMs Out of Self-Loops? Fine-Grained Reasoning Control with Activation Steering

arXiv:2607. 18100v1 Announce Type: new Abstract: Extended reasoning has become standard for frontier Large Language Models (LLMs), yet the trajectories these models produce remain largely uncontrollable.

By Sheldon Yu, Tong Yu, Xunyi Jiang, Rohan Surana, Gagan Mundada, Sungchul Kim, Lina Yao, Julian McAuley, Junda Wu

llms

Related stories

Is Your Model Thinking or Just Stagnating? PUMA: Diagnosing Reasoning Pathology via Phase-Momentum Alignment

CEDAR-GRPO: Process-Aware Reinforcement Learning for General Abductive Reasoning in LLMs

A Temporal Reasoning Benchmarking Framework for LRMs via Difficulty-controlled and Dynamic Test Generation

Divergent-Convergent Reasoning: Scaling Test-Time Compute through Structured Solution Synthesis

Process Reward Agents for Steering Knowledge-Intensive Reasoning

Can We Break LLMs Out of Self-Loops? Fine-Grained Reasoning Control with Activation Steering