Task-oriented Dialogue System任务型对话系统,其多用于垂直领域业务助理系统,如微软小娜、百度度秘、阿里小蜜以及我们研发的对话技术平台(DTP)等。这类系统具有明确需要完成的任务目标,如订餐、订票等。整个的任务型对话的框图里面,除了输入之外,除了语音信号和文本之外,比较大的三个模块就是自然语言理解,对话管理和自然语言生成。对话管理和语言生成的过程当中会遇到知识库和APIs。APIs在实际的应用过程中,可能会调一些查天气和地理位置的APIs,这些都可以被包含在任务型对话资源里面。
ACL, pp.270-278, (2020)
Neural response generation is a subcategory of text-generation that shares the objective of generating natural-looking text that is relevant to the prompt
Cited by154BibtexViews636DOI
0
0
ACM Transactions on Information Systems, no. 3 (2020): 1-32
Unlike task-oriented bots, most neural response generation models developed for open-domain dialog systems are not grounded in real world, which prevents these systems from effectively conversing about anything that relates to the user’s environment
Cited by39BibtexViews160DOI
0
0
Abhinav Rastogi, Xiaoxue Zang, Srinivas Sunkara, Raghav Gupta, Pranav Khaitan
national conference on artificial intelligence, (2020)
We introduced the schema-guided paradigm for taskoriented dialogue that simplifies the integration of new services and APIs with large scale virtual assistants
Cited by11BibtexViews58
0
0
ACL 2020, pp.85-96, (2020)
A novel pre-training model for dialogue generation is introduced in this paper, incorporated with latent discrete variables for one-to-many relationship modeling
Cited by4BibtexViews162
0
0
ACL, pp.5832-5841, (2020)
We propose Customized Model Agnostic Meta-Learning algorithm, which is able to customize unique dialogue models for different tasks
Cited by2BibtexViews624
0
0
Yinhe Zheng,Zikai Chen, Rongsheng Zhang, Shilei Huang,Xiaoxi Mao,Minlie Huang
An inverse dialogue model is introduced in our method to produce stylized pseudo dialogue pairs, which are used in a joint training process to train the stylized dialogue model
Cited by1BibtexViews85
0
0
ACL, pp.1417-1427, (2020)
We propose Perception Bot, a transmitter-receiver framework which explicitly models understanding between interlocutors
Cited by1BibtexViews187
0
0
ACL, pp.5821-5831, (2020)
We presented a three-stage framework, Generate-Delete-Rewrite, for persona consistent dialogue generation
Cited by0BibtexViews233
0
0
ACL, pp.6334-6343, (2020)
To induce the neural dialogue generation model to learn from more effective samples, we develop a gated data augmentation mechanism for the manipulation framework to selectively augment the learning samples
Cited by0BibtexViews110
0
0
Huda Khayrallah, João Sedoc
empirical methods in natural language processing, pp.4489-4505, (2020)
Simulated Multiple Reference Training improves upon a strong Transformer baseline in quality and diversity
Cited by0BibtexViews54DOI
0
0
Yan Zeng, Jian-Yun Nie
We investigated the general problem of conditioned dialogue generation
Cited by0BibtexViews53
0
0
EMNLP 2020, pp.904-916, (2020)
We showed existing dialogue agents are highly insensitive to contradiction, and introduced an orthogonally applicable method using the Rational Speech Acts framework to alleviate the issue
Cited by0BibtexViews64DOI
0
0
This paper presents the first topic-aware multi-turn dialogue modeling design in terms of explicitly segmenting and extracting topic-aware context utterances for retrieval-based dialogue systems
Cited by0BibtexViews75
0
0
Phillip Lippe, Pengjie Ren, Hinda Haned, Bart Voorn,Maarten de Rijke
We propose to combine the merits of templatebased Dialogue Response Generation and corpusbased Dialogue Response Generation in Task-oriented Dialogue Systems by presenting P2-Net based on prototype guided paraphrasing
Cited by0BibtexViews70
0
0
Science China-technological Sciences, no. 10 (2020): 2011-2027
We review the recent advancements on task-oriented dialog systems and discuss three critical topics: data efficiency, multi-turn dynamics, and knowledge integration
Cited by0BibtexViews216DOI
0
0
Pawel Budzianowski,Ivan Vulic
NGT@EMNLP-IJCNLP, pp.15-22, (2019)
We demonstrate that recent progress in language modeling pre-training and transfer learning shows promise to overcome this problem
Cited by29BibtexViews46DOI
0
0
IJCAI, (2019): 5190-5196
Experimental results show that our model outperforms the state-of-the-art methods, especially in terms of diversity and persona integration
Cited by16BibtexViews101DOI
0
0
Can Xu,Wei Wu,Chongyang Tao, Huang Hu, Matt Schuerman, Ying Wang
Meeting of the Association for Computational Linguistics, (2019)
We present a goal-tracking memory enhanced sequence-to-sequence model for open domain response generation with meta-words which explicitly define characteristics of responses
Cited by8BibtexViews123
0
0
meeting of the association for computational linguistics, (2018)
We focus on the standard dialogue task of predicting the utterance given the dialogue history, but consider this task both with and without the profile information being given to the learning agent
Cited by436BibtexViews295DOI
0
0
national conference on artificial intelligence, (2018)
We propose Emotional Chatting Machine that can generate appropriate responses in content and in emotion
Cited by376BibtexViews242
0
0