Question Answering on DROP
0.891ScoreALBERT
Evaluation Results
| Method | Links | |
|---|---|---|
| ALBERTseen_dataset_during_training=true2020.05 | 0.891 | |
| PerCEModel=Qwen3-4B2026.02 | 0.66 | |
| CEModel=Qwen3-4B2026.02 | 0.62 | |
| PerCEModel=Llama3-8B2026.02 | 0.612 | |
| CEModel=Llama3-8B2026.02 | 0.568 | |
| UnifiedQAseen_dataset_during_training=false, training_format=multi-format2020.05 | 0.325 | |
| UnifiedQA [AB]seen_dataset_during_training=false, training_format=abstractive2020.05 | 0.307 | |
| UnifiedQA [MC]seen_dataset_during_training=false, training_format=multiple choice2020.05 | 0.289 | |
| UnifiedQA [EX]seen_dataset_during_training=false, training_format=extractive2020.05 | 0.246 | |
| UnifiedQA [YN]seen_dataset_during_training=false, training_format=yes/no2020.05 | 0.004 |