Hybrid Generative-Retrieval Transformers for Dialogue Domain Adaptation
This addresses the challenge of adapting dialogue systems to new domains with limited data, which is crucial for real-world applications, though it is incremental as it builds on existing pre-training and fine-tuning methods.
The paper tackled the problem of few-shot domain adaptation in dialogue systems by introducing a hybrid generative-retrieval model based on GPT-2, achieving state-of-the-art results with over 4% improvement in human evaluation on the MetaLWOz dataset and competitive generalization on MultiWOZ.
Domain adaptation has recently become a key problem in dialogue systems research. Deep learning, while being the preferred technique for modeling such systems, works best given massive training data. However, in the real-world scenario, such resources aren't available for every new domain, so the ability to train with a few dialogue examples can be considered essential. Pre-training on large data sources and adapting to the target data has become the standard method for few-shot problems within the deep learning framework. In this paper, we present the winning entry at the fast domain adaptation task of DSTC8, a hybrid generative-retrieval model based on GPT-2 fine-tuned to the multi-domain MetaLWOz dataset. Robust and diverse in response generation, our model uses retrieval logic as a fallback, being SoTA on MetaLWOz in human evaluation (>4% improvement over the 2nd place system) and attaining competitive generalization performance in adaptation to the unseen MultiWOZ dataset.