Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens

ArXi:2508.01191v5 Announce Type: replace-cross Chain-of-Thought (CoT) prompting has been shown to be effective in eliciting structured reasoning (i.e., CoT reasoning) from large language models (LLMs). Regardless of its popularity, recent studies expose its failures in some reasoning tasks, raising fundamental questions about the nature of CoT reasoning. In this work, we propose a data distribution lens to understand when and why CoT reasoning succeeds or fails.