How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activation
Chain-of-Thought (CoT) prompting significantly enhances model reasoning, yet its internal mechanisms remain poorly understood. We analyze CoT's operational principles by reversely tracing information flow across decoding, projection, and activation phases. Our quantitative analysis suggests that CoT may serve as a decoding space pruner, leveraging answer templates to guide output generation, with higher template adherence strongly correlating with improved performance. Furthermore, we surprisingly find that CoT modulates neuron engagement in a task-dependent manner: reducing neuron activation in open-domain tasks, yet increasing it in closed-domain scenarios. These findings offer a novel mechanistic interpretability framework and critical insights for enabling targeted CoT interventions to design more efficient and robust prompts. We released our code and data at https://anonymous.4open.science/r/cot-D247.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Stream: Scaling up Mechanistic Interpretability to Long Context in LLMs via Sparse Attention
As Large Language Models (LLMs) scale to million-token contexts, traditional Mechanistic Interpretability techniques for analyzing attention scale quadratically with context length, demanding terabytes of memory beyond 1…
Detecting Unfaithful Chain-of-Thought via Circuit-Guided Internal-External Discrepancy
Chain-of-thought (CoT) reasoning improves the problem-solving ability of large language models (LLMs), but generated reasoning traces may not faithfully reflect the model's actual decision process. Existing CoT unfaithfu…
Pattern Analysis of Money Flow in the Bitcoin Blockchain
Bitcoin is the first and highest valued cryptocurrency that stores transactions in a publicly distributed ledger called the blockchain. Understanding the activity and behavior of Bitcoin actors is a crucial research topi…
Graph EmbeddingImaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language Models
Vision language models (VLMs) excel at many tasks but still struggle with spatial reasoning when critical information is not directly observable. Many such problems require imaginative perception: inferring what would be…
Spatial ReasoningSpectral-Progressive Thought Flow for Lightweight Multimodal Reasoning
Multimodal spatial reasoning often relies on long chains of intermediate textual and visual thoughts, where accumulating visual tokens and dense cross-modal attention incur substantial computation and memory overhead. To…
Multimodal ReasoningSpatial Reasoning