mlsanityData to 7 Oct 2026

Local Causal Attribution of Chain-of-Thought Reasoning

arXiv 2606.21821 ↗First posted 20 Jun 2026cs.LG cs.CLcited by 1

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

Abstract (arXiv)

Understanding the causal structure of a language model's thought process is a problem of significant importance for both transparency and safety. In this work, we take a local approach toward this goal by analyzing the causal relationships among individual components, termed units, of a given, specific chain-of-thought trace. We construct a structural causal model on these units and relate each unit to the log probability of generating (subsequent) output units. Our algorithm, termed AttriCoT, is a black-box method that performs attribution by estimating importance parameters in the structural causal model using O(U) forward passes through the model, where U is the number of units. Evaluation of perturbation curves across 5 datasets and 4 reasoning models shows that AttriCoT produces attributions that are more faithful to the model's behavior than alternative methods. The attribution results also reveal notable differences in thought structure between models and domains.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 7 of its 39 references that are in the map. First posted on arXiv 20 Jun 2026.

Builds on

7 of 39 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Is Chain-of-Thought Really Not Explainability? Chain-of-Thought Can Be Faithful without Hint Verbalization28 Dec
  2. 2025Reasoning Models Don't Always Say What They Think8 May
  3. 2023Measuring Faithfulness in Chain-of-Thought Reasoning17 Jul

Other references · 4

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  3. 2025Qwen3 Technical Reportcited by 8,195
  4. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026From Concept Alignment to Causal Grounding: An Intervention Test of Chain-of-Thought Faithfulnesscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Why think step by step? Reasoning emerges from the locality of experience
  2. 2026Dynamics Within Latent Chain-of-Thought: An Empirical Study of Causal Structure
  3. 2023Enhancing Causal Reasoning in Large Language Models: A Causal Attribution Model for Precision Fine-Tuning
  4. 2025Unveiling and Causalizing CoT: A Causal Pespective
  5. 2025FRIT: Using Causal Importance to Improve Chain-of-Thought Faithfulness
  6. 2025Rethinking the Chain-of-Thought: The Roles of In-Context Learning and Pre-trained Priors
  7. 2025Causal Sufficiency and Necessity Improves Chain-of-Thought Reasoning
  8. 2023Symbolic Chain-of-Thought Distillation: Small Models Can Also "Think" Step-by-Step