Towards Understanding the Cognitive Habits of Large Reasoning Models
Large Reasoning Models (LRMs), which autonomously produce a reasoning Chain of Thought (CoT) before producing final responses, offer a promising approach to interpreting and monitoring model behaviors. Inspired by the observation that certain CoT patterns -- e.g., ``Wait, did I miss anything?'' -- consistently emerge across tasks, we explore whether LRMs exhibit human-like cognitive habits. Building on Habits of Mind, a well-established framework of cognitive habits associated with successful human problem-solving, we introduce CogTest, a principled benchmark designed to evaluate LRMs' cognitive habits. CogTest includes 16 cognitive habits, each instantiated with 25 diverse tasks, and employs an evidence-first extraction method to ensure reliable habit identification. With CogTest, we conduct a comprehensive evaluation of 16 widely used LLMs (13 LRMs and 3 non-reasoning ones). Our findings reveal that LRMs, unlike conventional LLMs, not only exhibit human-like habits but also adaptively deploy them according to different tasks. Finer-grained analyses further uncover patterns of similarity and difference in LRMs' cognitive habit profiles, particularly certain inter-family similarity (e.g., Qwen-3 models and DeepSeek-R1). Extending the study to safety-related tasks, we observe that certain habits, such as Taking Responsible Risks, are strongly associated with the generation of harmful responses. These findings suggest that studying persistent behavioral patterns in LRMs' CoTs is a valuable step toward deeper understanding of LLM misbehavior. The code is available at: https://github.com/jianshuod/CogTest.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
End-to-End Graph Flattening Method for Large Language Models
In recent years, the breakthrough of Large Language Models (LLMs) offers new ideas for achieving universal methods on graph data. The common practice of converting graphs into natural language for LLMs, which refers to g…
See How You Read? Multi-Reading Habits Fusion Reasoning for Multi-modal Fake News Detection
The existing approaches based on different neural networks automatically capture and fuse the multimodal semantics of news, which have achieved great success for fake news detection. However, they still suffer from the l…
Fake News DetectionSurf or sleep? Understanding the influence of bedtime patterns on campus
Poor sleep habits may cause serious problems of mind and body, and it is a commonly observed issue for college students due to study workload as well as peer and social influence. Understanding its impact and identifying…
ManagementPatchCue: Enhancing Vision-Language Model Reasoning with Patch-Based Visual Cues
Vision-Language Models (VLMs) have achieved remarkable progress on a wide range of challenging multimodal understanding and reasoning tasks. However, existing reasoning paradigms, such as the classical Chain-of-Thought (…
Visual Question AnsweringReinforcement LearningVisual ReasoningUnderstanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory
While Large Reasoning Models (LRMs) generate extensive chain-of-thought reasoning, we lack a principled framework for understanding how these thoughts are structured. In this paper, we introduce a novel approach by apply…