mlsanityData to 7 Oct 2026

Analyzing Chain-of-Thought Prompting in Large Language Models via Gradient-based Feature Attributions

arXiv 2307.13339 ↗First posted 25 Jul 2023cs.CL cs.AIcited by 32

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

Abstract (arXiv)

Chain-of-thought (CoT) prompting has been shown to empirically improve the accuracy of large language models (LLMs) on various question answering tasks. While understanding why CoT prompting is effective is crucial to ensuring that this phenomenon is a consequence of desired model behavior, little work has addressed this; nonetheless, such an understanding is a critical prerequisite for responsible model deployment. We address this question by leveraging gradient-based feature attribution methods which produce saliency scores that capture the influence of input tokens on model output. Specifically, we probe several open-source LLMs to investigate whether CoT prompting affects the relative importances they assign to particular input tokens. Our results indicate that while CoT prompting does not increase the magnitude of saliency scores attributed to semantically relevant tokens in the prompt compared to standard few-shot prompting, it increases the robustness of saliency scores to question perturbations and variations in model output.

Key facts. Cited by 32 papers 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 10 of its 25 references that are in the map. First posted on arXiv 25 Jul 2023.

Builds on

10 of 25 references.

Closest ancestors · 4

Same direction, newest first

  1. 2022Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters20 Dec
  2. 2022Automatic Chain of Thought Prompting in Large Language Models7 Oct
  3. 2022Text and Patterns: For Effective Chain of Thought, It Takes Two to Tango16 Sep
  4. 2022Large Language Models are Zero-Shot Reasoners24 May

Other references · 6

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  3. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 895
  4. 2022Can language models learn from explanations in context?cited by 248
  5. 2022Interpreting Language Models with Contrastive Explanationscited by 91
  6. 2019AllenNLP Interpret: A Framework for Explaining Predictions of NLP Modelscited by 74

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 12; 2025: 12; 2026: 5.

32 papers.

Sort:

By area:

  1. 2023Explainability for Large Language Models: A Surveycited by 331
  2. 2023Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances, Frontiers and Futurecited by 184
  3. 2024The Impact of Reasoning Step Length on Large Language Modelscited by 152
  4. 2025Thought Anchors: Which LLM Reasoning Steps Matter?cited by 126
  5. 2023Unleashing the potential of prompt engineering for large language modelscited by 72
  6. 2025A Survey on Large Language Models for Mathematical Reasoningcited by 52
  7. 2024Explainable Artificial Intelligence: A Survey of Needs, Techniques, Applications, and Future Directioncited by 46
  8. 2025Transformers Learn to Implement Multi-step Gradient Descent with Chain of Thoughtcited by 32
  9. 2024Deciphering the Factors Influencing the Efficacy of Chain-of-Thought: Probability, Memorization, and Noisy Reasoningcited by 27
  10. 2024Unveiling the Statistical Foundations of Chain-of-Thought Prompting Methodscited by 27
  11. 2024Stop Reasoning! When Multimodal LLM with Chain-of-Thought Reasoning Meets Adversarial Imagecited by 26
  12. 2025Landscape of Thoughts: Visualizing the Reasoning Process of Large Language Modelscited by 25
20 more
  1. 2026Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Modelscited by 22
  2. 2024The Sound of Healthcare: Improving Medical Transcription ASR Accuracy with Large Language Modelscited by 17
  3. 2025Explainable Chain-of-Thought Reasoning: An Empirical Analysis on State-Aware Reasoning Dynamicscited by 16
  4. 2024A Theoretical Understanding of Chain-of-Thought: Coherent Reasoning and Error-Aware Demonstrationcited by 13
  5. 2024VLM's Eye Examination: Instruct and Inspect Visual Competency of Vision Language Modelscited by 8
  6. 2024Chain-of-Thought in Large Language Models: Decoding, Projection, and Activationcited by 6
  7. 2025Caught in the Act: a mechanistic approach to detecting deceptioncited by 6
  8. 2024Securing the Future of GenAI: Policy and Technologycited by 4
  9. 2026Do LLMs Encode Functional Importance of Reasoning Tokens?cited by 4
  10. 2024Towards detecting unanticipated bias in Large Language Modelscited by 3
  11. 2024Brain-like Functional Organization within Large Language Modelscited by 3
  12. 2025Which Words Matter Most in Zero-Shot Prompts?cited by 2
  13. 2025XiHeFusion: Harnessing Large Language Models for Science Communication in Nuclear Fusioncited by 2
  14. 2025Can Large Language Models Simulate Human Responses? A Case Study of Stated Preference Experiments in the Context of Heating-related Choicescited by 2
  15. 2026The Potential of CoT for Reasoning: A Closer Look at Trace Dynamicscited by 2
  16. 2026Reasoning Fails Where Step Flow Breakscited by 2
  17. 2025How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activationcited by 1
  18. 2025What Really Counts? Examining Step and Token Level Attribution in Multilingual CoT Reasoningcited by 1
  19. 2026Opening the Black Box: A Survey on the Mechanisms of Multi-Step Reasoning in Large Language Modelscited by 1
  20. 2025Revisiting the UID Hypothesis in LLM Reasoning Tracescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Triggering Chain-of-Thought via Latent Feature Interventions in Large Language Models
  2. 2026Attention-guided Fine-tuning of Multimodal Large Language Models Improves Chain-of-Thought Reasoning
  3. 2026Beyond the Prompt in Large Language Models: Comprehension, In-Context Learning, and Chain-of-Thought
  4. 2025Don't Take Things Out of Context: Attention Intervention for Enhancing Chain-of-Thought Reasoning in Large Language Models
  5. 2025How does Chain of Thought Think? Mechanistic Interpretability of Chain-of-Thought Reasoning with Sparse Autoencoding
  6. 2024Enhancing Chain of Thought Prompting in Large Language Models via Reasoning Patterns
  7. 2023An automatically discovered chain-of-thought prompt generalizes to novel models and datasets
  8. 2023Chain of Thought Prompt Tuning in Vision Language Models