Transformer-based models for transfer learning have the potential to achieve high prediction accuracies on text-based supervised learning tasks with relatively few training data instances. These mo...

中文翻译：

---

用于社会科学文本分析的 Transformer 神经迁移学习简介

基于 Transformer 的迁移学习模型有可能在训练数据实例相对较少的情况下，对基于文本的监督学习任务实现高预测精度。这些摩...