让大模型更聪明地问问题，通过概率规划提升对话推理成功率。
CA-BED: Conversation-Aware Bayesian Experimental Design
- 用贝叶斯方法动态规划提问，预测回答并评估信息价值。
- 在两个推理任务中，成功率比直接提问高21.8%。
- 只需多1.8轮对话，适合需要精准追问的交互场景。
大型语言模型在静态推理任务中表现优异，但在需主动提问获取信息的交互场景中性能常下降。核心挑战在于如何选择能降低不确定性的提问，同时处理模糊或不完整回答。为此，我们提出对话感知的贝叶斯实验设计（CA-BED），一种基于大模型似然估计的推理时概率对话规划框架。该方法维护假设的信念分布，模拟可能的回答，并通过对话树传播预期信息增益。在两个结构化实体推断基准上，CA-BED相比直接提示平均提升21.8%的成功率，与其它信息获取方法相当，且仅增加1.8轮对话。
原文摘要 · Abstract (English)
Large Language Models (LLMs) excel at static reasoning tasks, yet their performance often degrades in interactive scenarios where information must be actively acquired through questioning. A key challenge lies in selecting questions that reduce uncertainty while incorporating responses that may be ambiguous or only partially informative. To address this, we propose Conversation-Aware Bayesian Experimental Design (CA-BED), an inference-time probabilistic dialog planning framework that integrates Bayesian Experimental Design with LLM-based likelihood estimation to optimize question selection over multiple conversational turns. CA-BED maintains a belief distribution over hypotheses, anticipates possible answers, and propagates expected information gain through a simulated conversation tree. Across two structured entity-deduction benchmarks, CA-BED yields an average 21.8% improvement in success rates over direct prompting, with comparable gains relative to alternative information-seeking methods. It achieves these gains with an average increase of only 1.8 conversational turns compared to direct prompting.
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。