通过知识增强与对齐提升知识接地对话系统的事实一致性
Improving Factual Consistency for Knowledge-Grounded Dialogue Systems via Knowledge Enhancement and Alignment
AI总结：
针对知识接地对话易违背给定知识的问题，本文提出 K-Dial 扩展 FFN 并结合 RLFC 对齐事实偏好，在 WoW 与 CMU_DoG 上提升了回复事实一致性。
摘要：
基于预训练语言模型（PLM）的知识接地对话系统容易生成与所提供知识源在事实上不一致的回复。在此类不一致回复中，对话模型未能准确表达其所依赖的外部知识。受既往工作启发——该工作发现 Transformers 中的前馈网络（FFN）负责事实知识表达——我们研究了两种分别通过知识增强和对齐来高效提升 FFN 事实表达能力的方法。我们首先提出 K-Dial，它针对知识接地对话输入的特定模式，在 Transformers 中显式引入扩展 FFN，以增强事实知识表达。此外，我们采用面向事实一致性的强化学习（RLFC）方法，通过与黄金知识进行事实一致性偏好对齐，来隐式调整 FFN 在回复中的表达。为全面评估回复的事实一致性和对话质量，我们采用了大量自动指标和人工评估，包括复杂的、基于细粒度 NLI 的指标。在 WoW 和 CMU_DoG 数据集上的实验结果表明，我们的方法有效增强了 FFN 模块传达事实知识的能力，验证了在知识接地对话系统中改进事实一致性的效果。
英文摘要：
Pretrained language models (PLMs) based knowledge-grounded dialogue systems are prone to generate responses that are factually inconsistent with the provided knowledge source. In such inconsistent responses, the dialogue models fail to accurately express the external knowledge they rely upon. Inspired by previous work which identified that feed-forward networks (FFNs) within Transformers are responsible for factual knowledge expressions, we investigate two methods to efficiently improve the factual expression capability {of FFNs} by knowledge enhancement and alignment respectively. We first propose \textsc{K-Dial}, which {explicitly} introduces {extended FFNs in Transformers to enhance factual knowledge expressions} given the specific patterns of knowledge-grounded dialogue inputs. Additionally, we apply the reinforcement learning for factual consistency (RLFC) method to implicitly adjust FFNs' expressions in responses by aligning with gold knowledge for the factual consistency preference. To comprehensively assess the factual consistency and dialogue quality of responses, we employ extensive automatic measures and human evaluations including sophisticated fine-grained NLI-based metrics. Experimental results on WoW and CMU\_DoG datasets demonstrate that our methods efficiently enhance the ability of the FFN module to convey factual knowledge, validating the efficacy of improving factual consistency for knowledge-grounded dialogue systems.
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。
