UDALM: Unsupervised Domain Adaptation through Language Modeling
In this work we explore Unsupervised Domain Adaptation (UDA) of pretrained language models for downstream tasks. We introduce UDALM, a fine-tuning procedure, using a mixed classification and Masked Language Model loss, that can adapt to the target domain distribution in a robust and sample efficient manner. Our experiments show that performance of models trained with the mixed loss scales with the amount of available target data and the mixed loss can be effectively used as a stopping criterion during UDA training. Furthermore, we discuss the relationship between A-distance and the target error and explore some limitations of the Domain Adversarial Training approach. Our method is evaluated on twelve domain pairs of the Amazon Reviews Sentiment dataset, yielding $91.74\%$ accuracy, which is an $1.11\%$ absolute improvement over the state-of-the-art.
PDF Abstract NAACL 2021 PDF NAACL 2021 Abstract
Datasets
Results from the Paper
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
|---|---|---|---|---|---|---|
| Sentiment Analysis | Multi-Domain Sentiment Dataset | UDALM: Unsupervised Domain Adaptation through Language Modeling | DVD | 89.78 | # 1 | |
| Books | 90.63 | # 1 | ||||
| Electronics | 92.78 | # 1 | ||||
| Kitchen | 93.77 | # 1 | ||||
| Average | 91.74 | # 1 |
