DLGNet-Task: An End-to-end Neural Network Framework for Modeling Multi-turn Multi-domain Task-Oriented Dialogue

Abstract

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 opendomain dialogue systems. In this paper, we present a new framework, DLGNet-Task, a unified task-oriented dialogue system which employs autoregressive transformer networks such as DLGNet and GPT-2/3 to complete user tasks in multiturn multi-domain conversations. Our framework enjoys the controllable, verifiable, and explainable outputs of modular approaches, and the low development, deployment and maintenance cost of end-to-end systems. Treating open-domain system components as additional TOD system modules allows DLGNet-Task to learn the joint distribution of the inputs and outputs of all the functional blocks of existing modular approaches such as, natural language understanding (NLU), state tracking, action policy, as well as natural language generation (NLG). Rather than training the modules individually, as is common in real-world systems, we trained them jointly with appropriate module separations. When evaluated on the MultiWOZ2.1 dataset, DLGNet-Task shows comparable performance to the existing state-of-the-art approaches. Furthermore, using DLGNet-Task in conversational AI systems reduces the level of effort required for developing, deploying, and maintaining intelligent assistants at scale.

Towards Complex Scenarios: Building End-to-End Task-Oriented Dialogue System across Multiple Knowledge Bases

“…Existing EToDs can be classified into two main categories. The first strand of work (Ham et al 2020;Hosseini-Asl et al 2020;Olabiyi et al 2020;Peng et al 2021;Kulhánek et al 2021;Lee 2021;Yang, Li, and Quan 2021;Gao et al 2021;Gu et al 2021) treats all task-oriented dialogue pipeline tasks as a sequence prediction problem, using pre-trained models to predict dialog state, system action and system response in one sequence. These works can train all pipeline tasks in an end-to-end fashion but need annotations for intermediate results.…”