Published on Thu Sep 09 2021

Fusing task-oriented and open-domain dialogues in conversational agents

Tom Young, Frank Xing, Vlad Pandelea, Jinjie Ni, Erik Cambria

The goal of building intelligent dialogue systems has largely been pursued under two paradigms: task-oriented dialogue (TOD) systems and open-domain dialogue (ODD) systems. The two dialogue modes can potentially be intertwined together seamlessly in the same conversation. Such ability is desirable in conversational agents, as the integration makes them more accessible and useful.

0
0
0
Abstract

The goal of building intelligent dialogue systems has largely been \textit{separately} pursued under two paradigms: task-oriented dialogue (TOD) systems, which perform goal-oriented functions, and open-domain dialogue (ODD) systems, which focus on non-goal-oriented chitchat. The two dialogue modes can potentially be intertwined together seamlessly in the same conversation, as easily done by a friendly human assistant. Such ability is desirable in conversational agents, as the integration makes them more accessible and useful. Our paper addresses this problem of fusing TODs and ODDs in multi-turn dialogues. Based on the popular TOD dataset MultiWOZ, we build a new dataset FusedChat, by rewriting the existing TOD turns and adding new ODD turns. This procedure constructs conversation sessions containing exchanges from both dialogue modes. It features inter-mode contextual dependency, i.e., the dialogue turns from the two modes depend on each other. Rich dependency patterns including co-reference and ellipsis are features. The new dataset, with 60k new human-written ODD turns and 5k re-written TOD turns, offers a benchmark to test a dialogue model's ability to perform inter-mode conversations. This is a more challenging task since the model has to determine the appropriate dialogue mode and generate the response based on the inter-mode context. But such models would better mimic human-level conversation capabilities. We evaluate baseline models on this task, including \textit{classification-based} two-stage models and \textit{two-in-one} fused models. We publicly release FusedChat and the baselines to propel future work on inter-mode dialogue systems https://github.com/tomyoung903/FusedChat.

Wed Apr 29 2020
Machine Learning
UniConv: A Unified Conversational Neural Architecture for Multi-domain Task-oriented Dialogues
"UniConv" is a novel unified neural architecture for end-to-end conversational systems. It is designed to jointly train a State Tracker and Response Generator. It incorporates information from various input components and models dialogue acts and target responses.
0
0
0
Sat May 02 2020
NLP
A Simple Language Model for Task-Oriented Dialogue
SimpleTOD is a simple approach to task-oriented dialogue. It uses a single causal language model trained on all sub-tasks recast as a single sequence prediction problem. This allows SimpleTOD to fully leverage transfer learning from pre-trained, open domain, causal language models such
0
0
0
Sat Oct 17 2020
NLP
RiSAWOZ: A Large-Scale Multi-Domain Wizard-of-Oz Dataset with Rich Semantic Annotations for Task-Oriented Dialogue Modeling
RiSAWOZ is a large-scale multi-domain Chinese Wizard-of-Oz dataset. It contains 11.2K human-to-human (H2H) multi-turn semantically annotated dialogues. Each dialogue is labeled with comprehensive dialogue goal, domain and dialogue states.
0
0
0
Sun Oct 04 2020
Neural Networks
DLGNet-Task: An End-to-end Neural Network Framework for Modeling Multi-turn Multi-domain Task-Oriented Dialogue
Task oriented dialogue (TOD) requires the complex interleaving of a number of individually controllable components with strong guarantees for explainability and verifiability. This has made it difficult to adopt the multi-turn multi-domain dialogue generation capabilities of streamlined end-to-end systems.
0
0
0
Thu Feb 27 2020
NLP
CrossWOZ: A Large-Scale Chinese Cross-Domain Task-Oriented Dialogue Dataset
CrossWOZ is the first large-scale Chinese Cross-Domain Wizard-of-Oz task-oriented dataset. It contains 6K dialogue sessions and 102K utterances for 5 domains, including hotel, restaurant, attraction, metro, and taxi.
0
0
0
Thu Sep 05 2019
NLP
Towards Task-Oriented Dialogue in Mixed Domains
This work investigates the task-oriented dialogue problem in mixed-domain settings. We first show that a specialized state tracking component in multiple domains plays an important role. We then propose a hybrid system which is able to improve the belief tracking accuracy of about 28%.
0
0
0
Fri Apr 17 2020
NLP
Can You Put it All Together: Evaluating Conversational Agents' Ability to Blend Skills
A good open-domain conversational agent should be able to blend several qualities into one cohesive flow. We propose a new dataset, BlendedSkillTalk, to analyze how these capabilities would mesh together in a natural conversation. We also compare the performance of different training schemes.
1
8
24
Thu Oct 11 2018
NLP
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
BERT is designed to pre-train deep                bidirectional representations from unlabeled text. It can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks.
13
8
15
Mon Jan 27 2020
NLP
Towards a Human-like Open-Domain Chatbot
We present Meena, a multi-turn open-domain chatbot trained end-to-end on data mined and filtered from public domain social media conversations. This 2.6Bparameter neural network is simply trained to minimize perplexity of the next progressivelytoken. We also
3
1
3
Mon May 10 2021
NLP
Recent Advances in Deep Learning Based Dialogue Systems: A Systematic Survey
dialogue systems are a popular Natural Language Processing (NLP) task. In this survey, we mainly focus on the deep learning-based dialogue systems. We comprehensively review state-of-the-art research outcomes in dialogue systems and analyze them from two angles: model type and system type.
5
1
1
Thu Apr 01 2021
NLP
MultiWOZ 2.4: A Multi-Domain Task-Oriented Dialogue Dataset with Essential Annotation Corrections to Improve State Tracking Evaluation
The MultiWOZ 2.0 dataset was released in 2018. It consists of more than 10,000 task-oriented dialogues spanning 7 domains. There is substantial substantial noise in the state annotations, which hinders a proper evaluation of dialogue Tracking models.
1
0
1
Fri Jun 19 2015
NLP
A Neural Conversational Model
Conversational modeling is an important task in natural language understanding and machine intelligence. Previous approaches are often restricted to specific domains and require hand-crafted rules. We present a simple approach which uses the recently proposed sequence to sequence framework.
1
1
1