Neural Approaches to Conversational AI

By Jianfeng Gao, Michel Galley, Lihong Li. 2018

750 citations recorded by the source corpus.

This tutorial surveys neural approaches to conversational AI that were developed in the last few years. We group conversational systems into three categories: (1) question answering agents, (2) task-oriented dialogue agents, and (3) social bots. For each category, we present a review of state-of-the-art neural approaches, draw the connection between neural approaches and traditional symbolic approaches, and discuss the progress we have made and challenges we are facing, using specific systems and models as case studies.

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

Improving Multi-Task Deep Neural Networks via Knowledge Distillation for Natural Language Understanding (2019, neutral): “A recent survey is included in (Gao et al., 2019).”
Multi-Task Deep Neural Networks for Natural Language Understanding (2019, neutral): “A recent survey is included in Gao et al. (2018).”
ReCoRD: Bridging the Gap between Human and Machine Commonsense Reading Comprehension (2018, neutral): “See (Gao et al., 2018) for a recent survey.”
A Hybrid Neural Network Model for Commonsense Reasoning (2019, methodology): “In future work, we plan to extend HNN to more sophisticated reasoning tasks, especially those where large-scale language models like BERT and GPT do not perform well, as discussed in (Gao et al., 2019; Niven and Kao, 2019).”
Unified Language Model Pre-training for Natural Language Understanding and Generation (2019, neutral): “A recent survey is (Gao et al., 2019).”
Towards a Human-like Open-Domain Chatbot (2020, background): “Recent works also explore new frameworks such as adversarial learning (Li et al., 2017; Zhang et al., 2018c), variational autoencod-ing (Zhao et al., 2017; Gu et al., 2019), or both (Gao et al., 2019b) at the cost of added complexity and less scalability.”
Anticipating Safety Issues in E2E Conversational AI: Framework and Tooling (2021, background): “…“human-likeness” or “engagingness”, either by automatically comparing with a human ground-truth reference, e.g., by using similarity metrics such as BLEURT (Sellam et al., 2020) or BERTscore (Zhang et al., 2020a), or by asking humans to evaluate this manually (Deriu et al., 2020; Li et al., 2019).”
Does Gender Matter? Towards Fairness in Dialogue Systems (2019, background): “Dialogue systems have attracted increasing attention in the academic field (Chen et al., 2017; Gao et al., 2019) and have been widely”
Towards Coherent and Cohesive Long-form Text Generation (2018, neutral): “See Gao et al. (2018) for a comprehensive survey.”
BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage (2022, background): “Open-domain dialogue models While opendomain dialogue has a rich history (Chen et al., 2017; Gao et al., 2019; Ni et al., 2021) the area has made significant recent progress by pre-training ever-larger neural models.”
Survey of Hallucination in Natural Language Generation (2022, background): “A task-oriented dialogue system is often composed of several modules: a natural language understanding module, a dialogue manager, and an NLG module [40].”
CausaLM: Causal Model Explanation Through Counterfactual Language Models (2020, methodology): “In NLP, they were also shown to produce better language models, and are being widely used both for language representation and for classification in nearly every subfield (Tshitoyan et al. 2019; Gao et al. 2019; Lee et al. 2020; Feder et al. 2020).”
Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization (2018, background): “Neural conversational models are effective in generating coherent and relevant responses [1, 2, 3, 4, etc.].”
DIALOGPT : Large-Scale Generative Pre-training for Conversational Response Generation (2019, background): “Modelling conversations, however, presents distinct challenges in that human dialogue, which en-capsulates the possibly competing goals of two participants, is intrinsically more diverse in the range of potential responses (Li et al., 2016a; Zhang et al., 2018; Gao et al., 2019a,b).”
AlgaeDICE: Policy Gradient from Arbitrary Experience (2019, background): “…learning (RL) in conjunction with function approximation has proliferated in recent years, demonstrating successful applications in ﬁelds such as robotics (Andrychowicz et al., 2018; Nachum et al., 2019a), game playing (Mnih et al., 2013), and conversational systems (Gao et al., 2019).”
DualDICE: Behavior-Agnostic Estimation of Discounted Stationary Distribution Corrections (2019, background): “1 Introduction Reinforcement learning (RL) has recently demonstrated a number of successes in various domains, such as games [25], robotics [1], and conversational systems [11, 18].”
Reinforced Cross-Modal Matching and Self-Supervised Imitation Learning for Vision-Language Navigation (2018, background): “Exploration Much work has been done on improving exploration [4, 12, 16, 28, 36] because the trade-off between exploration and exploitation is one of the fundamental challenges in RL.”