超越承诺边界：探究大型推理模型中的附带思维链
Beyond the Commitment Boundary: Probing Epiphenomenal Chain-of-Thought in Large Reasoning Models
AI总结：
通过早期退出估计思维链步骤的因果重要性，发现推理中存在从瞬态猜测到稳定答案的“承诺边界”，后续步骤为附带现象，可提前退出以缩短推理长度达55%而不影响性能。
摘要：
思维链推理是语言模型推理时扩展的主导范式，但每个步骤对最终答案的因果影响尚不明确。我们通过早期退出估计每个步骤的因果重要性，并利用这一度量研究多个模型家族的推理轨迹中答案如何形成。在多种任务中，我们发现推理通常会跨越一个“承诺边界”——从瞬态中间猜测到稳定、高置信度答案的急剧转变。这种转变通常发生在单个步骤中，远在模型推理块结束之前，随后是“附带”的思维链步骤，这些步骤不改变最终答案概率。利用注意力探针，我们表明答案形成阶段可以从中间推理步骤中以高精度线性解码，并稳健地泛化到未见过的推理任务。我们利用这一信号在承诺边界处提前退出推理块，平均将思维链长度减少高达55%，而对模型性能影响微乎其微。
英文摘要：
Chain-of-thought (CoT) reasoning is the dominant paradigm for inference-time scaling in language models, yet the causal influence of individual steps on the final answer remains poorly understood. In this work, we use answer logits at the end of each reasoning step to estimate each step's causal importance to the final answer and intermediate guesses, shedding light on the answer formation process of several reasoning model families. Across diverse tasks, we find that reasoning typically crosses a commitment boundary, a sharp transition from transient intermediate guesses to a stable, high-confidence answer. This transition often happens in a single step, well before the model's reasoning block ends, and is followed by epiphenomenal CoT steps that leave the final answer probability unaltered. Using attention probes, we show that answer-formation stages can be linearly decoded from the activations of intermediate reasoning steps with high accuracy, showing robust generalization to unseen reasoning tasks. We leverage this property for early-exiting reasoning blocks at the commitment boundary location, reducing the length of CoTs up to 55% with negligible impact on model performance.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
