NLP Tasks on Consolidated NLP Tasks (eval)
48.44Score (Single Best Aug)Human-designed Augmentation
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Human-designed AugmentationTarget Model Size=large2024.04 | 48.44 | 53.32 | |
| LLM-based Augmentation (LLMDA)Target Model Size=medium2024.04 | 47.8 | 51.09 | |
| Human-designed AugmentationTarget Model Size=medium2024.04 | 47.25 | 49.52 | |
| Traditional AugmentationTarget Model Size=medium2024.04 | 46.92 | 49.72 | |
| LLM-based Augmentation (LLMDA)Target Model Size=large2024.04 | 45.88 | 52.17 | |
| Traditional AugmentationTarget Model Size=large2024.04 | 44.81 | 47.33 | |
| Human-designed AugmentationTarget Model Size=small2024.04 | 42.43 | 44.52 | |
| Traditional AugmentationTarget Model Size=small2024.04 | 42.37 | 44.18 | |
| LLM-based Augmentation (LLMDA)Target Model Size=small2024.04 | 42.03 | 48.47 |