Improving Open-Domain Dialogue Systems via Multi-Turn Incomplete Utterance Restoration

In multi-turn dialogue, utterances do not always take the full form of sentences. These incomplete utterances will greatly reduce the performance of open-domain dialogue systems. Restoring more incomplete utterances from context could potentially help the systems generate more relevant responses. To facilitate the study of incomplete utterance restoration for open-domain dialogue systems, a large-scale multi-turn dataset Restoration-200K is collected and manually labeled with the explicit relation between an utterance and its context. We also propose a “pick-and-combine” model to restore the incomplete utterance from its context. Experimental results demonstrate that the annotated dataset and the proposed approach significantly boost the response quality of both single-turn and multi-turn dialogue systems.

[1]  Paolo Dragone,et al.  Non-Sentential Utterances in Dialogue: Experiments in Classification and Interpretation , 2015, ArXiv.

[2]  Joelle Pineau,et al.  Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models , 2015, AAAI.

[3]  Percy Liang,et al.  Learning Symmetric Collaborative Dialogue Agents with Dynamic Knowledge Graph Embeddings , 2017, ACL.

[4]  Salim Roukos,et al.  Bleu: a Method for Automatic Evaluation of Machine Translation , 2002, ACL.

[5]  Jonathan Ginzburg,et al.  Classifying Non-Sentential Utterances in Dialogue: A Machine Learning Approach , 2007, CL.

[6]  Paolo Dragone,et al.  Classification and Resolution of Non-Sentential Utterances in Dialogue , 2016 .

[7]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[8]  Hai Zhao,et al.  Modeling Multi-turn Conversation with Deep Utterance Aggregation , 2018, COLING.

[9]  Ying Chen,et al.  Multi-Turn Response Selection for Chatbots with Deep Attention Matching Network , 2018, ACL.

[10]  Wei Bi,et al.  Learning to Abstract for Memory-augmented Conversational Response Generation , 2019, ACL.

[11]  Shuming Shi,et al.  Skeleton-to-Response: Dialogue Generation Guided by Retrieval Memory , 2018, NAACL.

[12]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[13]  Dongyan Zhao,et al.  Learning to Converse with Noisy Data: Generation with Calibration , 2018, IJCAI.

[14]  Vineet Kumar,et al.  Non-sentential Question Resolution using Sequence to Sequence Learning , 2016, COLING.

[15]  Chin-Yew Lin,et al.  ROUGE: A Package for Automatic Evaluation of Summaries , 2004, ACL 2004.

[16]  Jonathan Ginzburg,et al.  Using Machine Learning for Non-Sentential Utterance Classification , 2005, SIGDIAL.

[17]  Dongyan Zhao,et al.  How to Make Context More Useful? An Empirical Study on Context-Aware Neural Conversational Models , 2017, ACL.

[18]  Marjorie McShane,et al.  Detection and Resolution of Verb Phrase Ellipsis , 2016, LILT.

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

[20]  Jitendra Ajmera,et al.  A statistical approach for Non-Sentential Utterance Resolution for Interactive QA System , 2015, SIGDIAL Conference.

[21]  Vineet Kumar,et al.  Incomplete Follow-up Question Resolution using Retrieval based Sequence to Sequence Learning , 2017, SIGIR.

[22]  Jaime G. Carbonell,et al.  Discourse Pragmatics and Ellipsis Resolution in Task-Oriented Natural Language Interfaces , 1983, ACL.

[23]  Jackie Chi Kit Cheung,et al.  Verb Phrase Ellipsis Resolution Using Discriminative and Margin-Infused Algorithms , 2016, EMNLP.

[24]  Ronald G. Dreslinski,et al.  Sirius: An Open End-to-End Voice and Vision Personal Assistant and Its Implications for Future Warehouse Scale Computers , 2015, ASPLOS.

[25]  Zhoujun Li,et al.  Sequential Match Network: A New Architecture for Multi-turn Response Selection in Retrieval-based Chatbots , 2016, ArXiv.

[26]  Wei-Ying Ma,et al.  Hierarchical Recurrent Attention Network for Response Generation , 2017, AAAI.