mlsanityData to 7 Oct 2026

Chain-of-Thought in Large Language Models: Decoding, Projection, and Activation

arXiv 2412.03944 ↗First posted 5 Dec 2024cs.AIcited by 6

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

Abstract (arXiv)

Chain-of-Thought prompting has significantly enhanced the reasoning capabilities of large language models, with numerous studies exploring factors influencing its performance. However, the underlying mechanisms remain poorly understood. To further demystify the operational principles, this work examines three key aspects: decoding, projection, and activation, aiming to elucidate the changes that occur within models when employing Chainof-Thought. Our findings reveal that LLMs effectively imitate exemplar formats while integrating them with their understanding of the question, exhibiting fluctuations in token logits during generation but ultimately producing a more concentrated logits distribution, and activating a broader set of neurons in the final layers, indicating more extensive knowledge retrieval compared to standard prompts. Our code and data will be publicly avialable when the paper is accepted.

Key facts. Cited by 6 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 19 of its 49 references that are in the map. First posted on arXiv 5 Dec 2024.

Builds on

19 of 49 references.

Closest ancestors · 9

Same direction, newest first

  1. 2024Deciphering the Factors Influencing the Efficacy of Chain-of-Thought: Probability, Memorization, and Noisy Reasoning1 Jul
  2. 2024On the Representational Capacity of Neural Language Models with Chain-of-Thought Reasoning20 Jun
  3. 2024On the Hardness of Faithful Chain-of-Thought Reasoning in Large Language Models15 Jun
  4. 2024Chain-of-Thought Reasoning Without Prompting15 Feb
  5. 2024The Impact of Reasoning Step Length on Large Language Models10 Jan
  6. 2023Contrastive Chain-of-Thought Prompting15 Nov
  7. 2023Analyzing Chain-of-Thought Prompting in Large Language Models via Gradient-based Feature Attributions25 Jul
  8. 2023Invalid Logic, Equivalent Gains: The Bizarreness of Reasoning in Language Model Prompting20 Jul
  9. 2022Text and Patterns: For Effective Chain of Thought, It Takes Two to Tango16 Sep

Other references · 10

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  4. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,928
  5. 2022Language Models Are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thoughtcited by 334
  6. 2021Do Prompt-Based Models Really Understand the Meaning of their Prompts?cited by 324
  7. 2023The False Promise of Imitating Proprietary LLMscited by 254
  8. 2024CoCoT: Contrastive Chain-of-Thought Prompting for Large Multimodal Models with Multiple Image Inputscited by 56
  9. 2023Towards Understanding In-Context Learning with Contrastive Demonstrations and Saliency Mapscited by 35
  10. 2024An Investigation of Neuron Activation as a Unified Lens to Explain Chain-of-Thought Eliciting Arithmetic Reasoning of LLMscited by 20

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 2.

6 papers.

  1. 2026Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Modelscited by 22
  2. 2025Causal Sufficiency and Necessity Improves Chain-of-Thought Reasoningcited by 16
  3. 2025ChatMotion: A Multimodal Multi-Agent for Human Motion Analysiscited by 13
  4. 2025UORA: Uniform Orthogonal Reinitialization Adaptation in Parameter-Efficient Fine-Tuning of Large Modelscited by 11
  5. 2025SafeMT: Multi-turn Safety for Multimodal Language Modelscited by 4
  6. 2026Context Learning for Multi-Agent Discussioncited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2023Why Can Large Language Models Generate Correct Chain-of-Thoughts?
  2. 2024Chain-of-Thought Augmentation with Logit Contrast for Enhanced Reasoning in Language Models
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  4. 2025How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activation
  5. 2025Rethinking the Chain-of-Thought: The Roles of In-Context Learning and Pre-trained Priors
  6. 2024How to think step-by-step: A mechanistic understanding of chain-of-thought reasoning
  7. 2025How does Chain of Thought Think? Mechanistic Interpretability of Chain-of-Thought Reasoning with Sparse Autoencoding
  8. 2026Beyond the Prompt in Large Language Models: Comprehension, In-Context Learning, and Chain-of-Thought