MixText
2000년 도입 · 논문 3편에서 사용
MixText is a semi-supervised learning method for text classification, which uses a new data augmentation method called TMix. TMix creates a large amount of augmented training samples by interpolating text in hidden space. The technique leverages advances in data augmentation to guess low-entropy labels for unlabeled data, making them as easy to use as labeled data.
Text Augmentation · Natural Language ProcessingText Classification Models · Natural Language ProcessingSemi-Supervised Learning Methods · General