UniGen: Universal Domain Generalization for Sentiment Classification via Zero-shot Dataset Generation
This work addresses the real-world applicability issue for sentiment classification by enabling generalization to any domain sharing the label space, though it is incremental as it builds on existing dataset generation paradigms.
The paper tackles the problem of domain-specific limitations in dataset generation for sentiment classification by proposing a universal domain generalization approach that generates datasets regardless of the target domain, resulting in a method that achieves generalizability across various domains with a parameter set orders of magnitude smaller than pre-trained language models.
Although pre-trained language models have exhibited great flexibility and versatility with prompt-based few-shot learning, they suffer from the extensive parameter size and limited applicability for inference. Recent studies have suggested that PLMs be used as dataset generators and a tiny task-specific model be trained to achieve efficient inference. However, their applicability to various domains is limited because they tend to generate domain-specific datasets. In this work, we propose a novel approach to universal domain generalization that generates a dataset regardless of the target domain. This allows for generalization of the tiny task model to any domain that shares the label space, thus enhancing the real-world applicability of the dataset generation paradigm. Our experiments indicate that the proposed method accomplishes generalizability across various domains while using a parameter set that is orders of magnitude smaller than PLMs.