LEEETs-Dial：端到端任务导向对话系统中的语言趋同
LEEETs-Dial: Linguistic Entrainment in End-to-End Task-oriented Dialogue systems
AI总结：
针对端到端任务导向对话系统缺乏语言趋同机制的问题，本文在基于 GPT-2 的系统中利用共享词汇，提出训练实例加权、趋同专用损失和额外条件化三种方法，自动与人工评估均表明其显著提升了回复与用户的语言对齐效果。
摘要：
语言趋同（linguistic entrainment）或称对齐（alignment），是指对话参与者所使用的语言模式彼此趋于一致的现象。尽管趋同已被证明能够带来更自然的用户体验，但大多数对话系统并未对此作出任何安排。在本工作中，我们通过利用共享词汇，介绍了在基于 GPT-2 的端到端任务导向对话系统中实现对话趋同的方法。我们试验了训练实例加权、趋同专用损失以及额外条件化，以生成与用户对齐的回复。我们证明，正如自动评估指标和人工评估指标所共同确认的那样，与未经趋同优化的基础模型相比，所有三种方法都产生了显著更好的趋同效果。
英文摘要：
Linguistic entrainment, or alignment, represents a phenomenon where linguistic patterns employed by conversational participants converge to one another. While entrainment has been shown to produce a more natural user experience, most dialogue systems do not have any provisions for it. In this work, we introduce methods for achieving dialogue entrainment in a GPT-2-based end-to-end task-oriented dialogue system through the utilization of shared vocabulary. We experiment with training instance weighting, entrainment-specific loss, and additional conditioning to generate responses that align with the user. We demonstrate that all three approaches produce significantly better entrainment than the base, non-entrainment-optimized model, as confirmed by both automated and manual evaluation metrics.
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。
