Semantic Representation for Dialogue Modeling (2021-05-21T00:00:00.000000Z)

TL;DR

To the knowledge, this work is the first to leverage a formal semantic representation into neural dialogue modeling, and compares with the textual input, AMR explicitly provides core semantic knowledge and reduces data sparsity.

Abstract

Although neural models have achieved competitive results in dialogue systems, they have shown limited ability in representing core semantics, such as ignoring important entities. To this end, we exploit Abstract Meaning Representation (AMR) to help dialogue modeling. Compared with the textual input, AMR explicitly provides core semantic knowledge and reduces data sparsity. We develop an algorithm to construct dialogue-level AMR graphs from sentence-level AMRs and explore two ways to incorporate AMRs into dialogue systems. Experimental results on both dialogue understanding and response generation tasks show the superiority of our model. To our knowledge, we are the first to leverage a formal semantic representation into neural dialogue modeling.

Authors

Yulong Chen

6 papers

Xuefeng Bai

3 papers

Yue Zhang

7 papers

TL;DR

Abstract

Authors

References71 items

Representation

Improving BERT with Syntax-aware Local Attention

Investigating Typed Syntactic Dependencies for Targeted Sentiment Classification Using Graph Attention Neural Network

Conversational Semantic Parsing for Dialog State Tracking

Online Back-Parsing for AMR-to-Text Generation

Dialogue Relation Extraction with Document-Level Heterogeneous Graph Attention Networks

Do Syntax Trees Help Pre-trained Transformers Extract Information?

Line Graph Enhanced AMR-to-Text Generation with Mix-Order Graph Attention Networks

Structural Information Preserving for Graph-to-Text Generation

Reasoning with Latent Structure Refinement for Document-Level Relation Extraction

Dialogue-AMR: Abstract Meaning Representation for Dialogue

Dialogue-Based Relation Extraction

AMR Parsing via Graph-Sequence Iterative Inference

AvgOut: A Simple Output-Probability Measure to Eliminate Dull Responses

Improving Robustness of Task Oriented Dialog Systems

Leveraging Dependency Forest for Neural Medical Relation Extraction

Aspect-Level Sentiment Analysis Via Convolution over Dependency Tree

PLATO: Pre-trained Dialogue Generation Model with Discrete Latent Variable

Learning the Difference that Makes a Difference with Counterfactually-Augmented Data

GECOR: An End-to-End Generative Ellipsis and Co-reference Resolution Model for Task-Oriented Dialogue

Modeling Graph Structure in Transformer for Better AMR-to-Text Generation

Implicit Deep Latent Variable Models for Text Generation

Attention Guided Graph Convolutional Networks for Relation Extraction

AMR Parsing as Sequence-to-Graph Transduction

Improving Dialogue State Tracking by Discerning the Relevant Context

Semantic Neural Machine Translation Using AMR

Improving Robustness of Neural Dialog Systems in a Data-Efficient Way with Turn Dropout

Semantic Parsing for Task Oriented Dialog using Hierarchical Representations

Wizard of Wikipedia: Knowledge-Powered Conversational agents

MultiWOZ - A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling

Guided Neural Language Generation for Abstractive Summarization using Abstract Meaning Representation

CoQA: A Conversational Question Answering Challenge

QuAC: Question Answering in Context

AMR Beyond the Sentence: the Multi-sentence AMR corpus

Abstract Meaning Representation for Multi-Document Summarization

Toward Abstractive Summarization Using Semantic Representations

AMR Parsing as Graph Prediction with Latent Alignment

A Graph-to-Sequence Model for AMR-to-Text Generation

Hypothesis Only Baselines in Natural Language Inference

Linguistically-Informed Self-Attention for Semantic Role Labeling

Exploiting Semantics in Neural Machine Translation with Graph Convolutional Networks

DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset

Attention is All you Need

Neural AMR: Sequence-to-Sequence Models for Parsing and Generation

Doubly-Attentive Decoder for Multi-modal Neural Machine Translation

Layer Normalization

A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues

A Network-based End-to-End Trainable Task-oriented Dialogue System

A Diversity-Promoting Objective Function for Neural Conversation Models

Semantically Conditioned LSTM-based Natural Language Generation for Spoken Dialogue Systems

A Neural Conversational Model

Sequence to Sequence Learning with Neural Networks

Neural Machine Translation by Jointly Learning to Align and Translate

A Discriminative Graph-Based Parser for the Abstract Meaning Representation

Contextual domain classification in spoken language understanding systems using recurrent neural network

Abstract Meaning Representation for Sembanking

Data-Driven Response Generation in Social Media

Semantic Dialogue Modeling

A Survey of Decision Fusion and Feature Fusion Strategies for Pattern Classification

Gender Differences in Language Use: An Analysis of 14,000 Text Samples

The Proposition Bank: An Annotated Corpus of Semantic Roles

Bleu: a Method for Automatic Evaluation of Machine Translation

Centering: A Framework for Modeling the Local Coherence of Discourse

PEGASUS: A Spoken Language Interface for On-Line Air Travel Planning

On faithfulness and factu-ality in abstractive summarization

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Scaling Multi-Domain Dialogue State Tracking via Query Reformulation

Better Transition-Based AMR Parsing with a Refined Search Space

From TreeBank to PropBank

Orleans, Louisiana

Methods in Natural Language Processing

Field of Study

Journal Information

Name

Volume

Venue Information