Dual Latent Variable Model for Low-Resource Natural Language Generation in Dialogue Systems
This work addresses the challenge of low-resource natural language generation for dialogue systems, offering an incremental improvement over existing methods.
The paper tackles the problem of natural language generation in dialogue systems with limited labeled data by proposing a variational neural-based generation model that integrates variational inference and a novel auxiliary autoencoding training procedure. The model outperforms previous methods with sufficient data and maintains acceptable performance with scarce training data.
Recent deep learning models have shown improving results to natural language generation (NLG) irrespective of providing sufficient annotated data. However, a modest training data may harm such models performance. Thus, how to build a generator that can utilize as much of knowledge from a low-resource setting data is a crucial issue in NLG. This paper presents a variational neural-based generation model to tackle the NLG problem of having limited labeled dataset, in which we integrate a variational inference into an encoder-decoder generator and introduce a novel auxiliary autoencoding with an effective training procedure. Experiments showed that the proposed methods not only outperform the previous models when having sufficient training dataset but also show strong ability to work acceptably well when the training data is scarce.