mlsanityData to 14 Oct 2026

Understanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory

arXiv 2509.14662 ↗First posted 18 Sep 2025cs.AI cs.CL cs.LGcited by 9

On the map: LLM reasoning → Mathematical reasoning

Abstract (arXiv)

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 applying Schoenfeld's Episode Theory, a classic cognitive framework for human mathematical problem-solving, to analyze the reasoning traces of LRMs. We annotated thousands of sentences and paragraphs from model-generated solutions to math problems using seven cognitive labels (e.g., Plan, Implement, Verify). The result is the first publicly available benchmark for the fine-grained analysis of machine reasoning, including a large annotated corpus and detailed annotation guidebooks. Our preliminary analysis reveals distinct patterns in LRM reasoning, such as the transition dynamics between cognitive states. This framework provides a theoretically grounded methodology for interpreting LRM cognition and enables future work on more controllable and transparent reasoning systems.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Placed in Mathematical reasoning, one of 17 directions in LLM reasoning. Builds on 24 of its 44 references that are in the map. First posted on arXiv 18 Sep 2025.

Builds on

24 of 44 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025LIMO: Less is More for Reasoning5 Feb
  2. 2024Evaluating Mathematical Reasoning Beyond Accuracy8 Apr
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models5 Feb

Other references · 21

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  3. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  6. 2025s1: Simple test-time scalingcited by 1,341
  7. 2021Cross-Task Generalization via Natural Language Crowdsourcing Instructionscited by 699
  8. 2025Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRscited by 377
  9. 2024A Survey on Knowledge Distillation of Large Language Modelscited by 284
  10. 2025DeepSeek-R1 Thoughtology: Let's think about LLM Reasoningcited by 92
  11. 2025Self-rewarding correction for mathematical reasoningcited by 70
  12. 2025Missing Premise exacerbates Overthinking: Are Reasoning Models losing Critical Thinking Skill?cited by 58
9 more
  1. 2024CodeMind: Evaluating Large Language Models for Code Reasoningcited by 32
  2. 2024What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspectivecited by 29
  3. 2025Cognitive Memory in Large Language Modelscited by 29
  4. 2025Metacognition and Uncertainty Communication in Humans and Large Language Modelscited by 23
  5. 2025Trade-offs in Large Reasoning Models: An Empirical Analysis of Deliberative and Adaptive Reasoning over Foundational Capabilitiescited by 20
  6. 2023Gemini in Reasoning: Unveiling Commonsense in Multimodal Large Language Modelscited by 16
  7. 2024LLMs as Models for Analogical Reasoningcited by 16
  8. 2025What makes Reasoning Models Different? Follow the Reasoning Leader for Efficient Decodingcited by 8
  9. 2025Beyond Accuracy: The Role of Calibration in Self-Improving Large Language Modelscited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 5.

9 papers.

  1. 2025Can LLMs Estimate Student Struggles? Human-AI Difficulty Alignment with Proficiency Simulation for Item Difficulty Predictioncited by 7
  2. 2026Towards a Mechanistic Understanding of Large Reasoning Models: A Survey of Training, Inference, and Failurescited by 6
  3. 2026ReasonOps: Operator Segmentation for LLM Reasoning Tracescited by 5
  4. 2025ThinkPilot: Steering Reasoning Models via Automated Think-prefixes Optimizationcited by 3
  5. 2025Modeling Hierarchical Thinking in Large Reasoning Modelscited by 3
  6. 2025Soundness-Aware Level: A Microscopic Signature that Predicts LLM Reasoning Potentialcited by 1
  7. 2026Do LLMs Experience an Internal Polylogue? Investigating Reasoning through the Lens of Personascited by 1
  8. 2026Aligning Large Vision-Language Models at Test Time: A Trajectory-Guided Structured Sampling Approachcited by 1
  9. 2026Learning to Stop without Learning to Stop: Self-Supervised Confidence Training Improves Reasoning Efficiencycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Schoenfeld's Anatomy of Mathematical Reasoning by Language Models
  2. 2025Probing the "Psyche'' of Large Reasoning Models: Understanding Through a Human Lens
  3. 2025The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity
  4. 2025KisMATH: Do LLMs Have Knowledge of Implicit Structures in Mathematical Reasoning?
  5. 2026The Potential of CoT for Reasoning: A Closer Look at Trace Dynamics
  6. 2025Systematic Diagnosis of Brittle Reasoning in Large Language Models
  7. 2026Cognitive Profiling of LRMs' Reasoning Traces Using Bloom's Taxonomy
  8. 2026Unmasking Reasoning Processes: A Process-aware Benchmark for Evaluating Structural Mathematical Reasoning in LLMs