思维链中无全局规划：揭示大语言模型的潜在规划视界

摘要

本研究源于对思维链动态的先前互补性观察：大型语言模型在思维链显化前已展现出潜在规划能力，从而削弱了显式思维链的重要性；然而对于需要多步推理的任务，思维链仍具有关键作用。为深入理解LLM内部状态与其言语化推理轨迹之间的关系，我们通过探测方法Tele-Lens对跨任务域的隐藏状态进行研究，探究LLM的潜在规划能力。实证结果表明，LLM表现出短视特性，主要进行增量式状态转移而非精确的全局规划。基于此特性，我们提出一种增强思维链不确定性估计的假设，并验证了仅需少量关键思维链节点即可有效表征完整路径的不确定性。我们进一步强调了利用思维链动态特征的重要性，证明无需性能损失即可实现思维链捷径的自动识别。相关代码、数据及模型已发布于https://github.com/lxucs/tele-lens。

English

This work stems from prior complementary observations on the dynamics of Chain-of-Thought (CoT): Large Language Models (LLMs) is shown latent planning of subsequent reasoning prior to CoT emergence, thereby diminishing the significance of explicit CoT; whereas CoT remains critical for tasks requiring multi-step reasoning. To deepen the understanding between LLM's internal states and its verbalized reasoning trajectories, we investigate the latent planning strength of LLMs, through our probing method, Tele-Lens, applying to hidden states across diverse task domains. Our empirical results indicate that LLMs exhibit a myopic horizon, primarily conducting incremental transitions without precise global planning. Leveraging this characteristic, we propose a hypothesis on enhancing uncertainty estimation of CoT, which we validate that a small subset of CoT positions can effectively represent the uncertainty of the entire path. We further underscore the significance of exploiting CoT dynamics, and demonstrate that automatic recognition of CoT bypass can be achieved without performance degradation. Our code, data and models are released at https://github.com/lxucs/tele-lens.

思维链中无全局规划：揭示大语言模型的潜在规划视界

No Global Plan in Chain-of-Thought: Uncover the Latent Planning Horizon of LLMs

摘要

Support