← Back to all news
arXiv AI By Yuanhe Zhang, Ziwei Wang, Jie Ren, Haoran Gao, Zhenhong Zhou, Fanyu Meng, Cong Wu, Li Sun, Sen Su

When Reasoning Goes Astray: Attention Dynamics of Uncontrolled Reasoning

Read the original on arXiv AI →

The paper introduces RADAR, a method that models large reasoning models (LRMs) as four states and uses dynamic attention responses to detect uncontrolled reasoning in real time. RADAR identifies abnormal attention patterns that precede repetitive loops, and the authors demonstrate that realigning these patterns reduces looping while maintaining performance. The study offers a mechanistic explanation of how benign reasoning can degenerate into harmful behavior and provides actionable guidance for runtime interventions.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv AI
Jun 3

Thinking Past the Answer: Evaluating Harmful Overthinking in Large Reasoning Models

arXiv:2606. 02835v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) improve performance by generating explicit intermediate reasoning traces through increased test-time compute, yet the assumption that longer reasoning is consistently beneficial remains under-examined.

By Simone Caldarella, Davide Talon, Rahaf Aljundi, Elisa Ricci, Massimiliano Mancini
multimodalbenchmarks
More like this →
arXiv AI
Sep 3

When Can Large Reasoning Models Save Thinking? Mechanistic Analysis of Behavioral Divergence in Reasoning

The paper investigates why large reasoning models (LRMs) often continue to think even when prompted to stop, a phenomenon called "Still-thinking". By examining confidence at the thinking-termination boundary, internal attention divergences, and attention allocation across prompt segments, the authors find that high perplexity and greater attention to the original question correlate with continued thinking. They propose an attention‑intervention method that suppresses explicit reasoning, which reduces inefficiency but also lowers accuracy, underscoring a trade‑off between instruction compliance, inference speed, and correctness.

By Rongzhi Zhu, Yi Liu, Jiancheng Wang, Xiangyu Liu, Zequn Sun, Yiwei Wang, Yu Deng, Zijian Zhou, Wei Hu
More like this →
arXiv Machine Learning
Jun 2

Are Large Reasoning Models Interruptible?

arXiv:2510. 11713v4 Announce Type: replace-cross Abstract: Real-world applications of Large Reasoning Models (LRMs) often require reasoning about changing prompts or environments.

By Tsung-Han Wu, Mihran Miroyan, David M. Chan, Trevor Darrell, Narges Norouzi, Joseph E. Gonzalez
benchmarks
More like this →