A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues

By Iulian Serban, Alessandro Sordoni, Ryan J. Lowe, Laurent Charlin, Joelle Pineau, Aaron C. Courville, Yoshua Bengio. 2016

1,144 citations recorded by the source corpus.

Sequential data often possesses hierarchical structures with complex dependencies between sub-sequences, such as found between the utterances in a dialogue. To model these dependencies in a generative framework, we propose a neural network-based generative architecture, with stochastic latent variables that span a variable number of time steps. We apply the proposed model to the task of dialogue response generation and compare it with other recent neural-network architectures. We evaluate the model performance through a human evaluation study. The experiments demonstrate that our model improves upon recently proposed models and that the latent variables facilitate both the generation of meaningful, long and diverse responses and maintaining dialogue state.

Subfields: Artificial Intelligence, Natural Language 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 Natural Language Processing

Citations in the corpus

Scaling Neural Machine Translation (2018, methodology): “Similar sequence-to-sequence models are also applied to other natural language processing tasks, such as abstractive summarization (See et al., 2017; Paulus et al., 2018) and dialog (Sordoni et al., 2015; Serban et al., 2017; Dusek and Jurc´ıcek, 2016).”
Towards Exploiting Background Knowledge for Building Conversation Systems (2018, background): “…of large-scale dialog datasets (Lowe et al., 2015b; Ritter et al., 2010; Sordoni et al., 2015) leading to the development of several end-to-end conversation systems (Shang et al., 2015; Vinyals and Le, 2015; Li et al., 2016; Serban et al., 2016) which treat dialog as a sequence generation task.”
Does Gender Matter? Towards Fairness in Dialogue Systems (2019, background): “Dialogue systems are often built based on real human conversational data through machine learning especially deep learning techniques (Shang et al., 2015; Serban et al., 2016; Serban et al., 2017).”
Way Off-Policy Batch Deep Reinforcement Learning of Implicit Human Preferences in Dialog (2019, methodology): “[53, 52, 43, 20]), on different datasets (movie dialogs [9] and Reddit [20]).”
Towards Coherent and Engaging Spoken Dialog Response Generation Using Automatic Conversation Evaluators (2019, background): “Using maximum mutual information (MMI) (Li et al., 2015) was one of the earlier attempts to make conversational responses more diverse (Serban et al., 2016b,a).”
Ethical Challenges in Data-Driven Dialogue Systems (2017, methodology): “ected that these behaviours will be encoded in the underlying model. To this end, we run another experiment. We train two commonly used endto-end dialogue models, HRED (Serban et al. 2016) and VHRED (Serban et al. 2017c), on the Twitter dataset and sample responses using contexts from the validation”
Human-centric Dialog Training via Offline Reinforcement Learning (2020, methodology): “The underlying parameters of the VHRED model were as follows: Context RNN hidden size = 1000 , decoder hidden size = 1250 , encoder hidden size = 1250 , z embedding size = 600 , gradient clip = 1 .”
The challenge of realistic music generation: modelling raw audio at scale (2018, background): “Beyond raw audio generation, several generative models that capture large-scale structure have been proposed for images [2, 3, 12, 27], dialogue [40], 3D shapes [30] and symbolic music [37].”
Visual Dialog (2016, background): “Thus, similar to [46], as shown in Fig.”
Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization (2018, methodology): “For relevance evaluation, we adopt BLEU [38], ROUGE [39] and three embedding-based metrics following [8, 40].”
DIALOGPT : Large-Scale Generative Pre-training for Conversational Response Generation (2019, background): “Most open-domain neural response generation systems suffer from content or style inconsistency (Li et al., 2016b; Zhang et al., 2019; Gao et al., 2019c), lack of long-term contextual information (Serban et al., 2017), and blandness (Li et al., 2016a; Zhang et al., 2018; Qin et al., 2019).”
How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation (2016, background): “models are receiving increased attention, particularly using end-to-end training with neural networks (Serban et al., 2016; Sordoni et al., 2015; Vinyals and Le, 2015).”
Personalizing Dialogue Agents: I have a dog, do you have pets too? (2018, methodology): “For example, modern solutions that build an open-ended dialogue system to the Alexa challenge combine hand-coded and machine-learned elements (Serban et al., 2017a).”
Neural Approaches to Conversational AI (2018, positive): “The VHRED model incidentally helps reducing blandness as suggested by sample outputs of (Serban et al., 2017).”
TransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agents (2019, background): “Still, further inquiries in the capabilities of these neural network architectures and developments (Serban et al., 2016; Miao, Yu, and Blunsom, 2015; Sordoni et al., 2015; Serban et al., 2017; Li, Monroe, and Jurafsky, 2016; Li et al., 2017) indicated that they were limited which made…”
Towards an Automatic Turing Test: Learning to Evaluate Dialogue Responses (2017, methodology): “The VHRED model is an extension of the original hierarchical recurrent encoderdecoder (HRED) model (Serban et al., 2016a) with a turn-level stochastic latent variable.”