Paraphrase Augmented Task-Oriented Dialog Generation

By Silin Gao, Yichi Zhang, Zhijian Ou, Zhou Yu. 2020

92 citations recorded by the source corpus.

Neural generative models have achieved promising performance on dialog generation tasks if given a huge data set. However, the lack of high-quality dialog data and the expensive data annotation process greatly limit their application in real world settings. We propose a paraphrase augmented response generation (PARG) framework that jointly trains a paraphrase model and a response generation model to improve the dialog generation performance. We also design a method to automatically construct paraphrase training data set based on dialog state and dialog act labels. PARG is applicable to various dialog generation models, such as TSCP (Lei et al., 2018) and DAMD (Zhang et al., 2019). Experimental results show that the proposed framework improves these state-of-the-art dialog models further on CamRest676 and MultiWOZ. PARG also outperforms other data augmentation methods significantly in dialog generation tasks, especially under low resource settings.

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

A Simple Language Model for Task-Oriented Dialogue (2020, background): “For reference, oracle settings compare across a variety of settings against HDSA [9], ARDM [56], LaRL [63], PARG [16] can be found in the Supplementary Materials, but these comparisons are not essential for end-to-end contributions.”
Recent advances in deep learning based dialogue systems: a systematic survey (2021, methodology): “Gao et al. (2020c) solved the same problem in a conversational machine reading task.”
Paraphrase Generation: A Survey of the State of the Art (2021, background): “And it is also a good way for data augmentation (Kumar et al., 2019; Gao et al., 2020).”
AESOP: Paraphrase Generation with Adaptive Syntactic Control (2021, background): “On the one hand, paraphrase generation has beneﬁted a wide range of NLP applications, such as neural machine translation (Yang et al., 2019), dialogue generation (Gao et al., 2020), as well as improving model robustness (Huang et al., 2021) and interpretability (Jiang et al., 2019).”
Simulated Chats for Task-oriented Dialog: Learning to Generate Conversations from Instructions (2020, methodology): “…(Du and Black 2018) etc. Methods to generate paraphrases can vary - these include the use of syntactic parse trees to generate alternatives (Du and Black 2018), or generative models based on variational autoencoders (Malandrakis et al. 2019) and sequence to sequence models (Gao et al. 2020).”