Enhancing Dialog Coherence with Event Graph Grounded Content Planning

How to generate informative, coherent and sustainable open-domain conversations is a non-trivial task. Previous work on knowledge grounded conversation generation focus on improving dialog informativeness with little attention on dialog coherence. In this paper, to enhance multi-turn dialog coherence, we propose to leverage event chains to help determine a sketch of a multi-turn dialog. We first extract event chains from narrative texts and connect them as a graph. We then present a novel event graph grounded Reinforcement Learning (RL) framework. It conducts high-level response content (simply an event) planning by learning to walk over the graph, and then produces a response conditioned on the planned content. In particular, we devise a novel multi-policy decision making mechanism to foster a coherent dialog with both appropriate content ordering and high contextual relevance. Experimental results indicate the effectiveness of this framework in terms of dialog coherence and informativeness.

[1]  Alan Ritter,et al.  Data-Driven Response Generation in Social Media , 2011, EMNLP.

[2]  Hang Li,et al.  Neural Responding Machine for Short-Text Conversation , 2015, ACL.

[3]  Ramesh Nallapati,et al.  Labeled LDA: A supervised topic model for credit attribution in multi-labeled corpora , 2009, EMNLP.

[4]  Mitesh M. Khapra,et al.  Towards Exploiting Background Knowledge for Building Conversation Systems , 2018, EMNLP.

[5]  Zheng-Yu Niu,et al.  Knowledge Aware Conversation Generation with Explainable Reasoning over Augmented Graphs , 2019, EMNLP.

[6]  Seungwhan Moon,et al.  OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs , 2019, ACL.

[7]  Dongyan Zhao,et al.  Chat More If You Like: Dynamic Cue Words Planning to Flow Longer Conversations , 2018, ArXiv.

[8]  Jason Weston,et al.  Wizard of Wikipedia: Knowledge-Powered Conversational agents , 2018, ICLR.

[9]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[10]  Nathanael Chambers,et al.  Unsupervised Learning of Narrative Event Chains , 2008, ACL.

[11]  Ben Poole,et al.  Categorical Reparameterization with Gumbel-Softmax , 2016, ICLR.

[12]  Zheng-Yu Niu,et al.  Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation , 2020, AAAI.

[13]  Ming-Wei Chang,et al.  A Knowledge-Grounded Neural Conversation Model , 2017, AAAI.

[14]  Xiaoyan Zhu,et al.  Commonsense Knowledge Aware Conversation Generation with Graph Attention , 2018, IJCAI.

[15]  Ting Liu,et al.  Story Ending Prediction by Transferable BERT , 2019, IJCAI.

[16]  Jason Weston,et al.  Translating Embeddings for Modeling Multi-relational Data , 2013, NIPS.

[17]  Rudolf Kadlec,et al.  Improved Deep Learning Baselines for Ubuntu Corpus Dialogs , 2015, ArXiv.

[18]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[19]  Eric P. Xing,et al.  Target-Guided Open-Domain Conversation , 2019, ACL.

[20]  Nathanael Chambers,et al.  A Corpus and Cloze Evaluation for Deeper Understanding of Commonsense Stories , 2016, NAACL.

[21]  Xiaodong Liu,et al.  Conversing by Reading: Contentful Neural Conversation with On-demand Machine Reading , 2019, ACL.

[22]  Weinan Zhang,et al.  Exploring Implicit Feedback for Open Domain Conversation Generation , 2018, AAAI.

[23]  Jianfeng Gao,et al.  Deep Reinforcement Learning for Dialogue Generation , 2016, EMNLP.

[24]  Ting Liu,et al.  Constructing Narrative Event Evolutionary Graph for Script Event Prediction , 2018, IJCAI.

[25]  Hua Wu,et al.  Know More about Each Other: Evolving Dialogue Strategy via Compound Assessment , 2019, ACL.

[26]  Maxine Eskénazi,et al.  Rethinking Action Spaces for Reinforcement Learning in End-to-end Dialog Agents with Latent Variable Models , 2019, NAACL.

[27]  Jianfeng Gao,et al.  A Diversity-Promoting Objective Function for Neural Conversation Models , 2015, NAACL.