ReasoningFlow: Discourse Structures for Understanding LLM Reasoning Traces

• URL: http://arxiv.org/abs/2606.05402v1
• Date: Wed, 03 Jun 2026 20:12:26 GMT
• Title: ReasoningFlow: Discourse Structures for Understanding LLM Reasoning Traces
• Abstract summary: ReasoningFlow is a framework that captures the discourse structures of LRM reasoning traces into fine-grained directed acyclic graphs.<n>By analyzing ReasoningFlow graphs, we find that LRMs exhibit structurally similar traces, despite being trained from different base models.<n> ReasoningFlow reveals diverse fine-grained reasoning behaviors that can be used for better reasoning trace monitorability.
• Score: 13.927941324541356
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large reasoning models (LRMs) produce reasoning traces with non-linear structures, such as backtracking and self-correction, that complicate the evaluation and monitoring of the reasoning process. We introduce ReasoningFlow, a framework that captures the discourse structures of LRM reasoning traces into fine-grained directed acyclic graphs (DAGs). We develop and validate our annotation schema through careful manual annotation of 31 traces (2.1k steps), achieving high inter-annotator agreement, then scale to automatic annotation of 1,260 traces (247.7k steps) spanning three tasks (math, science, argumentation) and five models (Qwen2.5-32B-Inst, QwQ-32B, DeepSeek-V3, DeepSeek-R1, GPT-oss-120B). By analyzing ReasoningFlow graphs, we find: (1) LRMs exhibit structurally similar traces, despite being trained from different base models and potentially non-overlapping post-training data. (2) ReasoningFlow reveals diverse fine-grained reasoning behaviors (e.g., local verification, self-reflection, and assumptions) that can be used for better reasoning trace monitorability. (3) In LRMs, most of the erroneous steps are not used to derive final answers. (4) Mechanistic causal dependencies between steps do not reflect the language-level discourse structure. We release the dataset and code in: https://github.com/jinulee-v/reasoningflow.

Related papers

• PEARL: Auditable Repair for Scientific Reasoning Graph Extraction [4.931179793958283]
  We propose PEARL, a training-free framework that turns noisy graph responses into auditable reasoning graphs.<n>PEARL first materializes explicit graph content under a closed Peircean schema, then uses matched evidence-grounded judge feedback to repair rejected edge types, local inference steps, and terminal roots.<n>The graphs provide a reliability layer for research-agent and AI scientist that need inspectable reasoning traces rather than unconstrained graph regeneration.
  arXiv  Detail & Related papers  (2026-07-20T13:14:13Z)
• Shorthand for Thought: Compressing LLM Reasoning via Entropy-Guided Supertokens [4.0704009036918025]
  Reasoning in Large Language Models incurs significant inference-time compute.<n>We show that reasoning tokens split into two functional types: low-entropy textitstructural tokens (recurring phrases that scaffold the reasoning process) and higher-entropy textitorganic tokens (problem-specific content that drives toward a solution)<n>We derive textitsupertokens that capture frequent structural patterns, then teach the model to adopt them via supervised fine-tuning.
  arXiv  Detail & Related papers  (2026-04-29T07:06:43Z)
• On the Role of Reasoning Patterns in the Generalization Discrepancy of Long Chain-of-Thought Supervised Fine-Tuning [63.41902113656453]
  Supervised Fine-Tuning (SFT) on long Chain-of-Thought (CoT) trajectories has become a pivotal phase in building large reasoning models.<n>We conduct a comparative study using two sources of verified CoT trajectories generated by two competing models.<n>We find that SFT on textttDeepSeek-R1-0528 data achieves remarkably lower training loss, yet exhibits significantly worse generalization performance.
  arXiv  Detail & Related papers  (2026-04-02T07:00:54Z)
• Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement [66.51979814832332]
  model formulates procedural graph extraction as a multi-round reasoning process with dedicated structural and logical refinement.<n>Experiments demonstrate that model achieves substantial improvements in both structural correctness and logical consistency over strong baselines.
  arXiv  Detail & Related papers  (2026-01-27T04:00:48Z)
• Structured Reasoning for Large Language Models [59.215789462977206]
  We propose Structured Reasoning (SCR), a framework that decouples reasoning trajectories into explicit, evaluable, and trainable components.<n>SCR substantially improves reasoning efficiency and self-verification.<n>Compared with existing reasoning paradigms, it reduces output token length by up to 50%.
  arXiv  Detail & Related papers  (2026-01-12T04:04:01Z)
• Verifying Chain-of-Thought Reasoning via Its Computational Graph [23.32876195998818]
  Chain-of-Thought (CoT) verification methods predict correctness based on outputs (black-box) or activations (gray-box)<n>We introduce a white-box method: Circuit-based Reasoning Verification (CRV)<n>We show that attribution graphs of correct CoT steps, viewed as execution traces of the model's latent reasoning circuits, possess distinct structural fingerprints from those of incorrect steps.
  arXiv  Detail & Related papers  (2025-10-10T12:06:04Z)
• Do Cognitively Interpretable Reasoning Traces Improve LLM Performance? [15.970162088175593]
  Chain-of-Thought (CoT) traces generate intermediate reasoning traces before producing an answer.<n>A common but often implicit assumption is that CoT traces should be semantically meaningful and interpretable to the end user.<n>We investigate this question in the Open Book Question-Answering domain by supervised fine-tuning LLaMA and Qwen models.
  arXiv  Detail & Related papers  (2025-08-21T23:48:50Z)
• ReasoningFlow: Semantic Structure of Complex Reasoning Traces [9.328084104525834]
  ReasoningFlow parses traces into directed acyclic graphs, enabling the characterization of distinct reasoning patterns as subgraph structures.<n>This human-interpretable representation offers promising applications in understanding, evaluating, and enhancing the reasoning processes of LRMs.
  arXiv  Detail & Related papers  (2025-06-03T07:11:34Z)
• The Price of a Second Thought: On the Evaluation of Reasoning Efficiency in Large Language Models [54.88805865447848]
  We show that instruct models achieve higher efficiency overall, and problem difficulty affects efficiency.<n>We propose COTHINK, a simple two-stage pipeline: an instruct model drafts a brief outline, and a thinking model expands it.<n>On GSM8K, MATH500, and AIME24, COTHINK cuts token usage by 21.1% while keeping accuracy on four thinking models, and remains competitive with strong efficiency baselines.
  arXiv  Detail & Related papers  (2025-05-28T06:24:45Z)
• Beyond 'Aha!': Toward Systematic Meta-Abilities Alignment in Large Reasoning Models [86.88657425848547]
  Large reasoning models (LRMs) already possess a latent capacity for long chain-of-thought reasoning.<n>We explicitly align models with three meta-abilities: deduction, induction, and abduction, using automatically generated, self-verifiable tasks.<n>Our three stage-pipeline individual alignment, parameter-space merging, and domain-specific reinforcement learning, boosts performance by over 10% relative to instruction-tuned baselines.
  arXiv  Detail & Related papers  (2025-05-15T17:58:33Z)
• SEAL: Steerable Reasoning Calibration of Large Language Models for Free [58.931194824519935]
  Large Language Models (LLMs) have demonstrated compelling capabilities for complex reasoning tasks via the extended chain-of-thought (CoT) reasoning mechanism.<n>Recent studies reveal substantial redundancy in the CoT reasoning traces, which negatively impacts model performance.<n>We introduce SEAL, a training-free approach that seamlessly calibrates the CoT process, improving accuracy while demonstrating significant efficiency gains.
  arXiv  Detail & Related papers  (2025-04-07T02:42:07Z)
• Understanding Reasoning Ability of Language Models From the Perspective of Reasoning Paths Aggregation [110.71955853831707]
  We view LMs as deriving new conclusions by aggregating indirect reasoning paths seen at pre-training time. We formalize the reasoning paths as random walk paths on the knowledge/reasoning graphs. Experiments and analysis on multiple KG and CoT datasets reveal the effect of training on random walk paths.
  arXiv  Detail & Related papers  (2024-02-05T18:25:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.