Task-Oriented Dialogue System as Natural Language Generation

Abstract

In this paper, we propose to formulate the task-oriented dialogue system as the purely natural language generation task, so as to fully leverage the large-scale pre-trained models like GPT-2 and simplify complicated delexicalization prepossessing. However, directly applying this method heavily suffers from the dialogue entity inconsistency caused by the removal of delexicalized tokens, as well as the catastrophic forgetting problem of the pre-trained model during fine-tuning, leading to unsatisfactory performance. To alleviate these problems, we design a novel GPT-Adapter-CopyNet network, which incorporates the lightweight adapter and CopyNet modules into GPT-2 to achieve better performance on transfer learning and dialogue entity generation. Experimental results conducted on the DSTC8 Track 1 benchmark and MultiWOZ dataset demonstrate that our proposed approach significantly outperforms baseline models with a remarkable performance on automatic and human evaluations.

SIGIR 2022

References in corpus (16)

• HuggingFace's Transformers: State-of-the-art Natural Language Processing
• TransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agents
• A Simple Language Model for Task-Oriented Dialogue
• Parameter-Efficient Transfer Learning for NLP
• SOLOIST: Building Task Bots at Scale with Transfer Learning and Machine Teaching
• The Eighth Dialog System Technology Challenge
• Transferable Multi-Domain State Generator for Task-Oriented Dialogue Systems
• Incorporating BERT into Parallel Sequence Decoding with Adapters
• Alternating Recurrent Dialog Model with Large-scale Pre-trained Language Models
• ConvLab: Multi-Domain End-to-End Dialog System Platform
• UBAR: Towards Fully End-to-End Task-Oriented Dialog Systems with GPT-2
• Non-Autoregressive Dialog State Tracking
• Rethinking Action Spaces for Reinforcement Learning in End-to-end Dialog Agents with Latent Variable Models
• Teacher-Student Framework Enhanced Multi-domain Dialogue Generation
• Budgeted Policy Learning for Task-Oriented Dialogue Systems
• UniConv: A Unified Conversational Neural Architecture for Multi-domain Task-oriented Dialogues