Multi-view Response Selection for Human-Computer Conversation

By Xiangyang Zhou, Daxiang Dong, Hua Wu, Shiqi Zhao, Dianhai Yu, Hao Tian, Xuan Liu, Rui Yan. 2016

233 citations recorded by the source corpus.

In this paper, we study the task of response selection for multi-turn human-computer conversation. Previous approaches take word as a unit and view context and response as sequences of words. This kind of approaches do not explicitly take each utterance as a unit, therefore it is difﬁcult to catch utterance-level discourse information and dependencies. In this paper, we propose a multi-view response selection model that integrates information from two different views, i.e., word sequence view and utterance sequence view. We jointly model the two views via deep neural networks. Experimental results on a public corpus for context-sensitive response selection demonstrate the effectiveness of the proposed multi-view model, which signiﬁcantly outperforms other single-view baselines.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Deep Communicating Agents for Abstractive Summarization (2018, background): “…Lu-ong et al. (2015) proposes global and local attention networks for machine translation, while others investigate hierarchical attention networks for document classiﬁcation (Yang et al., 2016), sentiment classiﬁcation (Chen et al., 2016), and dialog response selection (Zhou et al., 2016).”
Challenges in Building Intelligent Open-domain Dialog Systems (2019, methodology): “[193] used a hierarchical Recurrent Neural Network (RNN) to encode a candidate and the utterance sequence in context, respectively, before computing their matching score.”
A Survey on Dialogue Systems: Recent Advances and New Frontiers (2017, background): “hen computed the matching degree score based on those two vectors. [97] selected the previous utterances in diﬀerent strategies and combined them with current messages to form a reformulated context. [111] performed context-response matching on not only the general word level context vector but also”
Response Generation by Context-aware Prototype Editing (2018, methodology): “Prior works on retrieval-based methods mainly focus on the matching model architecture for single turn conversation (Hu et al. 2014) and multi-turn conversation (Lowe et al. 2015; Zhou et al. 2016; Wu et al. 2017).”
Sequential Attention-based Network for Noetic End-to-End Response Selection (2019, methodology): “The typical approaches for multi-turn response selection mainly consist of sequence-based methods (Lowe et al. 2015; Yan, Song, and Wu 2016) and hierarchy-based meth-ods (Zhou et al. 2016; Wu et al. 2017; Zhang et al. 2018; Zhou et al. 2018).”
Sequential Match Network: A New Architecture for Multi-turn Response Selection in Retrieval-based Chatbots (2016, background): “(Zhou et al., 2016) improve multi-turn response selection with a multi-view model including an utterance view and a word view.”
Augmenting End-to-End Dialogue Systems With Commonsense Knowledge (2018, methodology): “Data-driven conversational models generally fall into two categories: retrieval-based methods (Lowe et al. 2015b; 2016a; Zhou et al. 2016), which select a response from a predefined repository, and generation-based methods (Ritter, Cherry, and Dolan 2011; Serban et al.”
Response Ranking with Deep Matching Networks and External Knowledge in Information-seeking Conversation Systems (2018, background): “ studies response ranking for single-turn conversation [35], which only considers a current utterance for selecting responses. Recently, several researchers have been studying multi-turn conversation [37, 40, 42, 49], which considers the previous utterances of the current message as the conversation”
Analyzing Assumptions in Conversation Disentanglement Research Through the Lens of a New Dataset and Model (2018, background): “…930,000 threads containing 100 million tokens, and has already been the basis of many papers (218 citations), particularly on developing dialogue agents (Hernandez et al., 2016; Zhou et al., 2016; Ouchi and Tsuboi, 2016; Serban et al., 2017; Wu et al., 2017; Zhang et al., 2018, inter alia).”
Multi-Turn Response Selection for Chatbots with Deep Attention Matching Network (2018, methodology): “Recent works extend it to multi-turn conversation scenario, Lowe et al.,(2015) and Zhou et al.,(2016) use RNN to read context and response, use the last hidden states to represent context and response as two semantic vectors, and measure their relevance.”
A Deep Look into Neural Ranking Models for Information Retrieval (2019, methodology): “Specifically, experiments have been conducted over neural ranking models based on collections such as Ubuntu Dialog Corpus (UDC) [75, 77, 78], Sina Weibo dataset [74, 26, 79, 80], MSDialog [81, 30, 82] and the ”campaign” NTCIR STC [83].”
Neural Response Generation via GAN with an Approximate Embedding Layer (2017, unknown): “Earlier works focused on paired word sequences only, while Zhou et al. (2016) and Iulian et al. (2017) have demonstrated that the comprehensibility of the generated responses can benefit from multiview training with respect to words, coarse tokens and utterances.”
Towards textual data augmentation for neural networks: synonyms and maximum loss (2019, methodology): “of sentence-level CNN embeddings [53], another model leveraging a memory network [11], and a different method using LSTMs to encode the message and response with the inner product for ranking [25].”
Enhance word representation for out-of-vocabulary on Ubuntu dialogue corpus (2018, methodology): “Zhou et al. (2016) treated each turn in multi-turn context as an unit and joined word sequence view and utterance sequence view together by deep neural networks.”
Recent Trends in Deep Learning Based Natural Language Processing [Review Article] (2018, methodology): “[159] SENTENCE-LEvEL CNN-LSTM ENCoDER 66.”
Domain Adaptive Training BERT for Response Selection (2019, methodology): “Previous studies of response selection concentrate on recurrent neural model to enhance dialog-response matching (Lowe et al., 2015; Zhou et al., 2016; Wu et al., 2017).”
Modeling Multi-turn Conversation with Deep Utterance Aggregation (2018, methodology): “Zhou et al. (2016) performed context-response matching with a multi-view model on both word level and utterance level.”