Natural Language Processing on 7 NLP Tasks Aggregate (T5-Base/Large Average)
88.9Accuracy (%)Fine-tuned
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Fine-tunedBackbone=T5-Large2025.05 | 88.9 | 169.1 | |
| NPSBackbone=T5-Large2025.05 | 88.8 | 39.2 | |
| TallS+TIESBackbone=T5-Large2025.05 | 88.3 | 54.3 | |
| Fine-tunedBackbone=T5-Base2025.05 | 83.1 | 47.8 | |
| NPSBackbone=T5-Base2025.05 | 82.9 | 11.1 | |
| TallS+TIESBackbone=T5-Base2025.05 | 82.6 | 15.2 | |
| TIESBackbone=T5-Large2025.05 | 80.3 | 25.1 | |
| Task ArithmeticBackbone=T5-Large2025.05 | 80.2 | 25.1 | |
| TIESBackbone=T5-Base2025.05 | 73.6 | 7.1 | |
| Task ArithmeticBackbone=T5-Base2025.05 | 73 | 7.1 | |
| Zero-shotBackbone=T5-Base2025.05 | 53.5 | 7.1 | |
| Zero-shotBackbone=T5-Large2025.05 | 53.1 | 25.1 |