How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activation 事件

How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activation arXiv:2507.20758v2 Announce Type: replace Abstract: 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 an