通过对比学习提升知识增强对话的鲁棒性
Improving the Robustness of Knowledge-Grounded Dialogue via Contrastive Learning
AI总结：
针对现实噪声影响知识增强对话（KGD）鲁棒性的问题，提出基于实体的对比学习框架，通过构造语义相关与无关的扰动正负样本提升模型抗噪能力，在三个基准数据集上取得最优性能且在噪声与少样本设置下表现更优。
摘要：
知识增强对话（KGD）学习基于给定的对话上下文和外部知识（例如，知识图谱；KGs）生成信息丰富的回复。最近，大型语言模型（LLMs）和预训练技术的出现为知识增强对话带来了巨大成功。然而，在实际应用中构建KGD系统时，会面临各种不可避免的现实世界噪声。例如，对话上下文可能涉及拼写错误和缩写等扰动。此外，KGs通常存在不完整问题，并且可能包含错误和过时的事实。这些现实世界噪声对KGD系统的鲁棒性提出了挑战，并阻碍了它们在现实世界中的应用。在本文中，我们提出了一种基于实体的对比学习框架来提升KGD的鲁棒性。具体而言，我们利用KGD样本中的实体信息来创建其正样本和负样本，这些样本分别涉及语义无关和语义相关的扰动。对比学习框架确保KGD模型能够感知这两类扰动，从而在真实应用中对潜在含噪输入生成信息丰富的回复。在三个基准数据集上的实验结果表明，我们的方法在自动评估分数上达到了新的最先进性能，验证了其有效性和潜力。此外，我们展示了在噪声和少样本设置下，我们的方法均能生成比对比模型更好的回复。
英文摘要：
Knowledge-grounded dialogue (KGD) learns to generate an informative response based on a given dialogue context and external knowledge (\emph{e.g.}, knowledge graphs; KGs). Recently, the emergence of large language models (LLMs) and pre-training techniques has brought great success to knowledge-grounded dialogue. However, when building KGD systems in real applications, there are various real-world noises that are inevitable to face. For example, the dialogue context might involve perturbations such as misspellings and abbreviations. In addition, KGs typically suffer from incompletion and also might contain erroneous and outdated facts. Such real-world noises pose a challenge to the robustness of KGD systems and hinder their applications in the real world. In this paper, we propose an entity-based contrastive learning framework for improving the robustness of KGD. Specifically, we make use of the entity information in a KGD sample to create both its positive and negative samples which involve semantic-irrelevant and semantic-relevant perturbations, respectively. The contrastive learning framework ensures the KGD model is aware of these two types of perturbations, thus generating informative responses with the potentially noisy inputs in real applications. Experimental results on three benchmark datasets show that our method achieves new state-of-the-art performance in terms of automatic evaluation scores, verifying its effectiveness and potentiality. Furthermore, we show that our method can generate better responses than comparison models in both the noisy and the few-shot settings.
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。
