arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2110.12949cs.CLcs.LG
收藏

通过生成对话进行角色认证

Persona Authentication through Generative Dialogue

Michigan State University(密歇根州立大学);Weill Cornell Medicine(威尔康奈尔医学院)

Fengyi Tang, Lifan Zeng, Fei Wang, Jiayu Zhou

摘要:

在本文中,我们定义和调查的问题\emph{persona authentication}:学习会话策略,以验证人物模型的一致性。我们提出了一个学习目标,并证明(在一些温和的假设下),在此目标下训练的局部密度估计最大化的人物角色信息和对话轨迹之间的互信息。基于所提出的目标,我们开发了一种学习认证模型的方法,该认证模型自适应地输出个性化的问题,以揭示其合作伙伴在整个多回合会话过程中的潜在角色。实验表明,我们的认证方法发现有效的问题序列,推广到看不见的人物配置文件。

英文摘要:

In this paper we define and investigate the problem of \emph{persona authentication}: learning a conversational policy to verify the consistency of persona models. We propose a learning objective and prove (under some mild assumptions) that local density estimators trained under this objective maximize the mutual information between persona information and dialog trajectory. Based on the proposed objective, we develop a method of learning an authentication model that adaptively outputs personalized questions to reveal the underlying persona of its partner throughout the course of multi-turn conversation. Experiments show that our authentication method discovers effective question sequences that generalize to unseen persona profiles.

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。

↑