mlsanityData to 14 Oct 2026

What Makes a Good Reasoning Chain? Uncovering Structural Patterns in Long Chain-of-Thought Reasoning

arXiv 2505.22148 ↗First posted 28 May 2025cs.AIcited by 21

On the map: LLM reasoning → Prompting and chain-of-thought reasoning

Abstract (arXiv)

Recent advances in reasoning with large language models (LLMs) have popularized Long Chain-of-Thought (LCoT), a strategy that encourages deliberate and step-by-step reasoning before producing a final answer. While LCoTs have enabled expert-level performance in complex tasks, how the internal structures of their reasoning chains drive, or even predict, the correctness of final answers remains a critical yet underexplored question. In this work, we present LCoT2Tree, an automated framework that converts sequential LCoTs into hierarchical tree structures and thus enables deeper structural analysis of LLM reasoning. Using graph neural networks (GNNs), we reveal that structural patterns extracted by LCoT2Tree, including exploration, backtracking, and verification, serve as stronger predictors of final performance across a wide range of tasks and models. Leveraging an explainability technique, we further identify critical thought patterns such as over-branching that account for failures. Beyond diagnostic insights, the structural patterns by LCoT2Tree support practical applications, including improving Best-of-N decoding effectiveness. Overall, our results underscore the critical role of internal structures of reasoning chains, positioning LCoT2Tree as a powerful tool for diagnosing, interpreting, and improving reasoning in LLMs.

Key facts. Cited by 21 papers in the map (data through 14 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 23 of its 50 references that are in the map. First posted on arXiv 28 May 2025.

Builds on

23 of 50 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models12 Mar
  2. 2025Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?26 Feb
  3. 2025LLMs Can Easily Learn to Reason from Demonstrations Structure, not content, is what matters!11 Feb
  4. 2025When More is Less: Understanding Chain-of-Thought Length in LLMs11 Feb

Other references · 19

  1. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  2. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
  3. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  4. 2025s1: Simple test-time scalingcited by 1,341
  5. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  6. 2024Large Language Monkeys: Scaling Inference Compute with Repeated Samplingcited by 991
  7. 2024DeepSeek-V3 Technical Reportcited by 811
  8. 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMscited by 570
  9. 2025LIMO: Less is More for Reasoningcited by 498
  10. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 415
  11. 2025Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRscited by 377
  12. 2024Skywork-Reward: Bag of Tricks for Reward Modeling in LLMscited by 327
7 more
  1. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 320
  2. 2024Free Process Rewards without Process Labelscited by 165
  3. 2025Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMscited by 163
  4. 2025Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learningcited by 161
  5. 2025The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Taskscited by 152
  6. 2025The relationship between reasoning and performance in large language models--o3 (mini) thinks harder, not longercited by 41
  7. 2025R-PRM: Reasoning-Driven Process Reward Modelingcited by 38

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 9; 2026: 12.

21 papers.

Sort:

By area:

  1. 2026Reinforcing Chain-of-Thought Reasoning with Self-Evolving Rubricscited by 26
  2. 2025What Characterizes Effective Reasoning? Revisiting Length, Review, and Structure of CoTcited by 25
  3. 2025Towards Concise and Adaptive Thinking in Large Reasoning Models: A Surveycited by 18
  4. 2025From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligencecited by 15
  5. 2026Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignmentcited by 13
  6. 2025First Try Matters: Revisiting the Role of Reflection in Reasoning Modelscited by 12
  7. 2025Lexical Hints of Accuracy in LLM Reasoning Chainscited by 9
  8. 2025The Signal is in the Steps: Local Scoring for Reasoning Data Selectioncited by 8
  9. 2025Schoenfeld's Anatomy of Mathematical Reasoning by Language Modelscited by 8
  10. 2026Beyond SFT-to-RL: Pre-alignment via Black-Box On-Policy Distillation for Multimodal RLcited by 8
  11. 2025Humans Perceive Wrong Narratives from AI Reasoning Textscited by 6
  12. 2026Towards a Mechanistic Understanding of Large Reasoning Models: A Survey of Training, Inference, and Failurescited by 6
9 more
  1. 2025Your thoughts tell who you are: Characterize the reasoning patterns of LRMscited by 4
  2. 2026BODHI: Do LLMs Branch Out and Discover Heterogeneous Inferences?cited by 2
  3. 2026Janus-Q: End-to-End Event-Driven Trading via Hierarchical-Gated Reward Modelingcited by 1
  4. 2026Strategy Executability in Mathematical Reasoning: Leveraging Human-Model Differences for Effective Guidancecited by 1
  5. 2026Reforming the Mechanism: Editing Reasoning Patterns in LLMs with Circuit Reshapingcited by 1
  6. 2026Do LLMs Experience an Internal Polylogue? Investigating Reasoning through the Lens of Personascited by 1
  7. 2026STOP: Structured On-Policy Pruning of Long-Form Reasoning in Low-Data Regimescited by 1
  8. 2026TRACES: Tagging Reasoning Steps for Adaptive Cost-Efficient Early-Stoppingcited by 0
  9. 2026Reasoning-Preserving Fine-Tuning of Post-RL LLMs with Null-Basis LoRAcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026LCoT-GV: Graph Attention Networks for Verifying Long Reasoning Chains in Large Language Models
  2. 2025Understanding Chain-of-Thought in Large Language Models via Topological Data Analysis
  3. 2025Chain or tree? Re-evaluating complex reasoning from the perspective of a matrix of thought
  4. 2026Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Models
  5. 2025Reasoning Beyond Language: A Comprehensive Survey on Latent Chain-of-Thought Reasoning
  6. 2026ChainPrune: Evaluating and Reducing Redundancy in Long Chain-of-Thought Reasoning
  7. 2025Multi-chain Graph Refinement and Selection for Reliable Reasoning in Large Language Models
  8. 2026Mean-Field Dynamics of Chain-of-Thought Reasoning in Large Language Models