Incorporating external knowledge into dialogue generation has been proven to benefit the performance of an open-domain Dialogue System (DS), such as generating informative or stylized responses, controlling conversation topics. In this article, we study the open-domain DS that uses unstructured text as external knowledge sources (\textbf{U}nstructured \textbf{T}ext \textbf{E}nhanced \textbf{D}ialogue \textbf{S}ystem, \textbf{UTEDS}). The existence of unstructured text entails distinctions between UTEDS and traditional data-driven DS and we aim to analyze these differences. We first give the definition of the UTEDS related concepts, then summarize the recently released datasets and models. We categorize UTEDS into Retrieval and Generative models and introduce them from the perspective of model components. The retrieval models consist of Fusion, Matching, and Ranking modules, while the generative models comprise Dialogue and Knowledge Encoding, Knowledge Selection, and Response Generation modules. We further summarize the evaluation methods utilized in UTEDS and analyze the current models' performance. At last, we discuss the future development trends of UTEDS, hoping to inspire new research in this field.
Wangxuan Institute of Computer Technology, Peking University(北京大学王选计算机技术研究所);Center for Data Science, AAIS, Peking University(北京大学人工智能研究院数据科学中心);Microsoft Corporation(微软公司);Beijing Academy of Artificial Intelligence (BAAI)(北京人工智能研究院)
clem:todd: A Framework for the Systematic Benchmarking of LLM-Based Task-Oriented Dialogue System Realisations
Computational Linguistics, Department of Linguistics University of Potsdam(普腾多夫大学语言学系);German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心)
Re-evaluating ADEM: A Deeper Look at Scoring Dialogue Responses
Indian Institute of Technology, Madras(印度理工学院马德拉斯分校);Robert Bosch Center for Data Sciences and AI (RBC-DSAI), Indian Institute of Technology, Madras(罗伯特博世数据科学与人工智能中心(RBC-DSAI),印度理工学院马德拉斯分校);Microsoft, India(微软印度公司);NVIDIA, India(英伟达印度公司)
自动评估非结构化领域的对话响应的质量是一个具有挑战性的问题。ADEM(Lowe et al. 2017)将对话系统的自动评估表述为一个学习问题,并表明这种模型能够预测与人类判断显著相关的反应,无论是在话语还是系统层面。他们的系统被证明击败了单词重叠指标,如BLEU,具有很大的利润率。我们从对手是否能利用ADEM模型的问题开始。针对基于神经网络的ADEM评估系统设计了一系列有针对性的攻击,表明对话系统的自动评估还有很长的路要走。ADEM可能会与文本中的词序颠倒这样简单的变体混淆!我们报告几个这样的对抗性的情况下,得出违反直觉的分数的对话反应的实验。我们采取了系统的看ADEM提出的评分功能,并将其连接到线性系统理论来预测系统中明显的缺点。我们还设计了一种攻击,可以欺骗这样的系统,以评价响应生成系统作为有利的。最后,我们提到未来的研究方向,使用对抗攻击,设计一个真正的自动化对话评估系统。
Manual-Guided Dialogue for Flexible Conversational Agents
Tsinghua University(清华大学);Tencent Inc.(腾讯公司);Beijing National Research Center for Information Science and Technology(北京信息科学与技术国家研究中心);WeChat AI, Tencent Inc.(腾讯微信人工智能部)
Bootstrapping incremental dialogue systems from minimal data: the generalisation power of dialogue grammars
Heriot-Watt University(赫瑞瓦特大学)
我们研究了一种端到端的方法,用于从少量未注释的对话数据自动诱导基于任务的对话系统。它结合了增量语义语法-动态语义和类型论(DS-TTR)-与强化学习(RL),其中语言生成和对话管理是一个联合决策问题。这样产生的系统是渐进的:对话是逐字处理的,以前被证明是支持自然的,自发的对话必不可少的。我们假设,语法中丰富的语言知识应该能够处理大量的对话变体,即使在很少的对话上进行训练。我们的实验表明,我们的模型可以处理74%的Facebook AI bAbI数据集,即使只训练了0.13%的数据(5个对话)。它还可以处理65%的bAbI+,这是我们通过系统地添加增量对话现象(如重启和自我纠正)创建的语料库。我们将我们的模型与最先进的检索模型MemN 2N进行比较。我们发现,在语义准确性方面,MemN 2N对bAbI+转换的鲁棒性非常差,即使在完整的bAbI数据集上进行训练也是如此。
Information Sciences Institute, University of Southern California(南加州大学信息科学研究所);Computer Science Department, Columbia University(哥伦比亚大学计算机科学系);Computer Science Department, University of California, Los Angeles(加州大学洛杉矶分校计算机科学系);Computer Science Department, University of California, Davis(加州大学戴维斯分校计算机科学系)
SpokenUS: A Spoken User Simulator for Task-Oriented Dialogue
Graduate School of Data Science, Seoul National University(首尔国立大学数据科学研究生院);Department of Information Systems, Hanyang University(翰阳大学信息系统系);Department of Computer Science and Engineering, Seoul National University(首尔国立大学计算机科学与工程系)
Bridging Reasoning and Action: Hybrid LLM-RL Framework for Efficient Cross-Domain Task-Oriented Dialogue
School of Computer Science and Technology, Changsha University of Science and Technology(长沙理工大学计算机科学与技术学院);School of Computer Science and Engineering, Central South University(中南大学计算机科学与工程学院);Institute of Computing and Intelligence, Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳)计算与智能研究院);University of Science and Technology Beijing(北京科技大学);Text Computing and Cognitive Intelligence MOE Engineering Research Center, Guizhou University(贵州大学文字计算与智能认知MOE工程研究中心)
Can Neural Generators for Dialogue Learn Sentence Planning and Discourse Structuring?
University of California, Santa Cruz(加州大学圣克鲁兹分校)
任务型对话系统中的应答往往实现多个命题,其最终形式取决于句子规划和语篇构建操作的使用。例如,推荐可以由明确的评价话语组成,例如Chanpen Thai是最佳选项,以及由理由话语关系相关的内容,例如It has great food and service,其将多个命题组合成单个短语。虽然神经生成方法在一个端到端学习框架中集成了句子规划和表面实现,但以前的工作尚未表明神经生成器可以:(1)执行常见的句子规划和话语结构化操作;(2)决定是在单个句子中还是在多个句子中实现内容;(3)决定是否在一个句子中实现内容。(3)概括训练中所见之外的句子规划和语篇关系操作。我们系统地创建大型训练语料库,展示特定的句子规划操作,然后测试神经模型,看看它们学到了什么。我们比较了没有显式潜变量的句子规划模型与在训练过程中提供显式监督的模型。我们表明,只有额外的监督模型可以重现句子规划和话语操作,并推广到训练中看不到的情况。
A Survey of the Evolution of Language Model-Based Dialogue Systems: Data, Task and Models
The Chinese University of Hong Kong(香港中文大学);Huawei Noah Ark Lab(华为诺亚实验室);MoE Key Laboratory of High Confidence Software Technologies, The Chinese University of Hong Kong(高可信软件技术重点实验室,香港中文大学)
WHEN TO ACT, WHEN TO WAIT: Modeling the Intent-Action Alignment Problem in Dialogue
Northeastern University(东北大学);Tufts University(塔夫茨大学);Boston University(波士顿大学);University of Texas at San Antonio(德克萨斯大学圣安东尼奥分校);Massachusetts Institute of Technology(麻省理工学院);Northwestern University(西北大学);George Washington University(乔治华盛顿大学)
DiaCDM: Cognitive Diagnosis in Teacher-Student Dialogues using the Initiation-Response-Evaluation Framework
Lab of Artificial Intelligence for Education, East China Normal University, China(人工智能教育实验室,华东师范大学,中国);Shanghai Institute of Artificial Intelligence for Education, East China Normal University, China(人工智能教育上海研究院,华东师范大学,中国);XR System Application Research Center, Nanchang Research Institute, Zhejiang University, China(XR系统应用研究中心,南昌研究所,浙江大学,中国)