用概念链提升大模型日常对话能力，效果优于现有提示方法。
Chain-of-Conceptual-Thought Elicits Daily Conversation in Large Language Models
- 先生成情感、策略、主题等概念，再展开内容，分步引导对话
- 在内外域概念场景下均超越自洽、ECoT等基线方法
- 适合需要自然流畅对话的场景，如情感支持和日常交流
思维链（CoT）广泛用于提升大模型在数学、编程和推理任务中的表现，但在开放域任务中受限于缺乏明确的推理步骤或逻辑过渡。为此，我们提出一种新的基于提示的范式——概念链（CoCT），要求大模型先生成概念标签（包括情感、策略和主题），再据此完成具体内容。为鼓励这种分层思考方式，我们设计了多维度概念体系。我们在日常对话和情感支持任务中测试该范式，涵盖域内与域外概念设置。自动评估、人工评价及基于大模型的评估均表明，CoCT显著优于自洽（self-refine）、ECoT、SoT和RAG等多种基线方法，展现出在更广泛任务中应用大模型提示范式的潜力。
原文摘要 · Abstract (English)
Chain-of-Thought (CoT) is widely applied to enhance the LLM capability in math, coding and reasoning tasks. However, its performance is limited for open-domain tasks, when there are no clearly defined reasoning steps or logical transitions. To mitigate such challenges, we propose a new prompt-based paradigm called Chain of Conceptual Thoughts (CoCT), which suggests the LLM first to produce the tag of concepts, then complete the detailed content following the concept. To encourage this hierarchical way of thinking, we implement the concepts with emotions, strategies and topics. We experiment with this paradigm in daily and emotional support conversations, covering tasks with both in-domain and out-of-domain concept settings. Automatic, human, and LLM-based evaluations reveal that CoCT surpasses several prompt-based baselines such as self-refine, ECoT, SoT and RAG, suggesting a potential solution of LLM prompting paradigm for a wider scope of tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。