{"ID":2876883,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.00190","arxiv_id":"2509.00190","title":"Explainable Chain-of-Thought Reasoning: An Empirical Analysis on State-Aware Reasoning Dynamics","abstract":"Recent advances in chain-of-thought (CoT) prompting have enabled large language models (LLMs) to perform multi-step reasoning. However, the explainability of such reasoning remains limited, with prior work primarily focusing on local token-level attribution, such that the high-level semantic roles of reasoning steps and their transitions remain underexplored. In this paper, we introduce a state-aware transition framework that abstracts CoT trajectories into structured latent dynamics. Specifically, to capture the evolving semantics of CoT reasoning, each reasoning step is represented via spectral analysis of token-level embeddings and clustered into semantically coherent latent states. To characterize the global structure of reasoning, we model their progression as a Markov chain, yielding a structured and interpretable view of the reasoning process. This abstraction supports a range of analyses, including semantic role identification, temporal pattern visualization, and consistency evaluation.","short_abstract":"Recent advances in chain-of-thought (CoT) prompting have enabled large language models (LLMs) to perform multi-step reasoning. However, the explainability of such reasoning remains limited, with prior work primarily focusing on local token-level attribution, such that the high-level semantic roles of reasoning steps an...","url_abs":"https://arxiv.org/abs/2509.00190","url_pdf":"https://arxiv.org/pdf/2509.00190v2","authors":"[\"Sheldon Yu\",\"Yuxin Xiong\",\"Junda Wu\",\"Xintong Li\",\"Tong Yu\",\"Xiang Chen\",\"Ritwik Sinha\",\"Jingbo Shang\",\"Julian McAuley\"]","published":"2025-08-29T18:53:31Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}