ThoughtProbe：通过探测表征实现分类器引导的 LLM 思维空间探索
ThoughtProbe: Classifier-Guided LLM Thought Space Exploration via Probing Representations
AI总结：
本文提出 ThoughtProbe 推理时框架，利用 LLM 隐藏特征作为判别信号引导树状响应空间探索，通过分类器评分优先扩展高分节点并聚合分支 CoT 分数，在多个算术推理基准上显著提升性能。
摘要：
本文介绍了 ThoughtProbe，这是一种新颖的推理时框架，它利用大型语言模型（LLMs）的隐藏推理特征来提升其推理性能。与以往操纵隐藏表征以引导 LLM 生成的工作不同，我们将其作为判别信号来引导树状结构响应空间的探索。在每个节点扩展中，分类器作为评分和排序机制，通过优先考虑得分较高的候选者进行延续，从而高效分配计算资源。在完成树扩展后，我们从所有分支收集答案以形成候选答案池。然后，我们提出了一种分支聚合方法，通过聚合所有支持分支的 CoT 分数对其进行边际化处理，从而从池中识别出最佳答案。实验结果表明，我们框架的全面探索不仅覆盖了有效的推理链，还能有效识别它们，在多个算术推理基准上取得了显著提升。
英文摘要：
This paper introduces ThoughtProbe, a novel inference time framework that leverages the hidden reasoning features of Large Language Models (LLMs) to improve their reasoning performance. Unlike previous works that manipulate the hidden representations to steer LLM generation, we harness them as discriminative signals to guide the tree structured response space exploration. In each node expansion, a classifier serves as a scoring and ranking mechanism that efficiently allocates computational resources by prioritizing higher score candidates for continuation. After completing the tree expansion, we collect answers from all branches to form a candidate answer pool. We then propose a branch aggregation method that marginalizes over all supporting branches by aggregating their CoT scores, thereby identifying the optimal answer from the pool. Experimental results show that our framework's comprehensive exploration not only covers valid reasoning chains but also effectively identifies them, achieving significant improvements across multiple arithmetic reasoning benchmarks.
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。
