通过表征工程解锁大语言模型的通用长思维链推理能力
Unlocking General Long Chain-of-Thought Reasoning Capabilities of Large Language Models via Representation Engineering
AI总结：
该研究从表征角度验证长 CoT 推理是 LLM 的通用且区别于普通 CoT 的能力，并提出表征工程方法 GLoRE，在领域内与跨领域场景中高效提升推理表现。
摘要：
近期，长思维链（long CoTs）的进展显著提升了大语言模型（LLMs）的推理能力。已有工作发现，仅通过少量样本进行调优，即可有效激发长 CoT 推理能力，并且该能力能够轻松迁移到其他任务。这促使我们研究长 CoT 推理是否是 LLMs 的一种通用能力。在本工作中，我们从表征角度对这一问题进行实证分析。我们发现，LLMs 确实将长 CoT 推理编码为一种通用能力，并且其与普通 CoTs 存在明显区别。此外，要实现长 CoT 推理的有效迁移，还需要特定领域的表征。受这些发现启发，我们提出 GLoRE，这是一种新颖的表征工程方法，用于释放 LLMs 的通用长 CoT 推理能力。大量实验证明了 GLoRE 在领域内和跨领域场景中的有效性与效率。
英文摘要：
Recent advancements in long chain-of-thoughts(long CoTs) have significantly improved the reasoning capabilities of large language models(LLMs). Existing work finds that the capability of long CoT reasoning can be efficiently elicited by tuning on only a few examples and can easily transfer to other tasks. This motivates us to investigate whether long CoT reasoning is a general capability for LLMs. In this work, we conduct an empirical analysis for this question from the perspective of representation. We find that LLMs do encode long CoT reasoning as a general capability, with a clear distinction from vanilla CoTs. Furthermore, domain-specific representations are also required for the effective transfer of long CoT reasoning. Inspired by these findings, we propose GLoRE, a novel representation engineering method to unleash the general long CoT reasoning capabilities of LLMs. Extensive experiments demonstrate the effectiveness and efficiency of GLoRE in both in-domain and cross-domain scenarios.
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。
