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.
Source: MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text ClassificationPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Classification | 1 | 25.00% |
General Classification | 1 | 25.00% |
Semi-Supervised Text Classification | 1 | 25.00% |
Text Classification | 1 | 25.00% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |