三思而后言：为响应生成显式生成隐式常识知识
Think Before You Speak: Explicitly Generating Implicit Commonsense Knowledge for Response Generation
AI总结：
本文提出Think-Before-Speaking（TBS）方法，通过先外化隐式常识知识再生成响应，提升对话系统的信息量和可解释性，在多数指标上超越基线。
摘要：
诸如常识等隐式知识是流畅人类对话的关键。当前的神经响应生成（RG）模型被训练为直接生成响应，忽略了未言明的隐式知识。在本文中，我们提出了Think-Before-Speaking（TBS），这是一种生成式方法，首先将隐式常识知识外化（思考），并利用该知识生成响应（发言）。我们期望外化隐式知识能实现更高效的学习，产生更具信息量的响应，并使模型更具可解释性。我们分析了收集知识对齐对话、表示隐式知识以及在知识与对话间进行转换的不同选择。实证结果表明，TBS模型在大多数自动指标上优于端到端和知识增强的RG基线，并且根据人类标注者的评估，能生成更具信息量、更具体且更遵循常识的响应。TBS生成的知识在约85%的情况下是合理且与对话相关的。
英文摘要：
Implicit knowledge, such as common sense, is key to fluid human conversations. Current neural response generation (RG) models are trained to generate responses directly, omitting unstated implicit knowledge. In this paper, we present Think-Before-Speaking (TBS), a generative approach to first externalize implicit commonsense knowledge (think) and use this knowledge to generate responses (speak). We expect that externalizing implicit knowledge allows more efficient learning, produces more informative responses, and enables more explainable models. We analyze different choices to collect knowledge-aligned dialogues, represent implicit knowledge, and transition between knowledge and dialogues. Empirical results show TBS models outperform end-to-end and knowledge-augmented RG baselines on most automatic metrics and generate more informative, specific, and commonsense-following responses, as evaluated by human annotators. TBS also generates knowledge that makes sense and is relevant to the dialogue around 85\% of the time.
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。
