| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Machine Comprehension | CBT-CN (test) | Accuracy83.7 | 56 | |
| Machine Comprehension | CBT NE (test) | Accuracy81.6 | 56 | |
| Machine Comprehension | CBT-CN (val) | Accuracy85.7 | 37 | |
| Machine Comprehension | CBT (test) | Named Entities73.2 | 12 | |
| CBT Conversation Generation | CBT conversation evaluation dataset | Semantic Coherence1.94 | 10 | |
| Information Exchange | CBT | F1 Score72.2 | 10 | |
| Zero-shot Language Modeling | CBT (test) | Accuracy84.2 | 4 |