# ReasoningFlow: Discourse Structures for Understanding LLM Reasoning Traces

arXiv: https://arxiv.org/abs/2606.05402 · HTML: https://mlsanity.com/paper/2606.05402/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 29 of its 77 references that are in the map.
- First posted on arXiv 3 Jun 2026.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 29 of 29)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [Solving math word problems with process- and outcome-based feedback](https://mlsanity.com/paper/2211.14275/) (2022)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Reasoning Models Don't Always Say What They Think](https://mlsanity.com/paper/2505.05410/) (2025)
- [Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs](https://mlsanity.com/paper/2503.01307/) (2025)
- [Demystifying Long Chain-of-Thought Reasoning in LLMs](https://mlsanity.com/paper/2502.03373/) (2025)
- [OpenThoughts: Data Recipes for Reasoning Models](https://mlsanity.com/paper/2506.04178/) (2025)
- [Imitate, Explore, and Self-Improve: A Reproduction Report on Slow-thinking Reasoning Systems](https://mlsanity.com/paper/2412.09413/) (2024)
- [Thought Anchors: Which LLM Reasoning Steps Matter?](https://mlsanity.com/paper/2506.19143/) (2025)
- [Olmo 3](https://mlsanity.com/paper/2512.13961/) (2025)
- [ReasonFlux-PRM: Trajectory-Aware PRMs for Long Chain-of-Thought Reasoning in LLMs](https://mlsanity.com/paper/2506.18896/) (2025)
- [Emergent Hierarchical Reasoning in LLMs through Reinforcement Learning](https://mlsanity.com/paper/2509.03646/) (2025)
- [BudgetThinker: Empowering Budget-aware LLM Reasoning with Control Tokens](https://mlsanity.com/paper/2508.17196/) (2025)
- [Cognitive Foundations for Reasoning and Their Manifestation in LLMs](https://mlsanity.com/paper/2511.16660/) (2025)
- [Dissociation of Faithful and Unfaithful Reasoning in LLMs](https://mlsanity.com/paper/2405.15092/) (2024)
- [Lost at the Beginning of Reasoning](https://mlsanity.com/paper/2506.22058/) (2025)
- [Can Aha Moments Be Fake? Towards Quantifying Decorative and True Thinking in Chain-of-Thought](https://mlsanity.com/paper/2510.24941/) (2025)
- [Scaling Evaluation-time Compute with Reasoning Models as Evaluators](https://mlsanity.com/paper/2503.19877/) (2025)
- [ReasoningFlow: Semantic Structure of Complex Reasoning Traces](https://mlsanity.com/paper/2506.02532/) (2025)
- [Resisting Contextual Interference in RAG via Parametric-Knowledge Reinforcement](https://mlsanity.com/paper/2506.05154/) (2025)
- [ReJump: A Tree-Jump Representation for Analyzing and Improving LLM Reasoning](https://mlsanity.com/paper/2512.00831/) (2025)
- [Towards Robust Process Reward Modeling via Noise-aware Learning](https://mlsanity.com/paper/2601.12748/) (2026)
- [From Chains to DAGs: Probing the Graph Structure of Reasoning in LLMs](https://mlsanity.com/paper/2601.17593/) (2026)

## Built on by (most cited first, 5 of 5)

- [ReasoningLens: Hierarchical Visualization and Diagnostic Auditing for Large Reasoning Models](https://mlsanity.com/paper/2606.23404/) (2026)
- [Beneath the Surface of Chains-of-Thought: A Mechanistic Interpretation of Reasoning Operations in LLMs](https://mlsanity.com/paper/2609.04753/) (2026)
- [From Concentration to Differentiation and Back: Routing Effective Rank in MoE Reasoning Cohorts](https://mlsanity.com/paper/2609.06403/) (2026)
- [SEABench: Benchmarking Endogenous Misalignment In Self-Evolving Agents](https://mlsanity.com/paper/2609.35596/) (2026)
- [Learning Process Rewards via Reasoning State Propagation](https://mlsanity.com/paper/2609.39220/) (2026)

## Abstract (arXiv)

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.