CANARD
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Question Rewriting | CANARD (test) | Hard Difficulty Score0.4133 | 13 | |
| Conversational Question Answering | CANARD | F1 Score0.681 | 9 | |
| Conversational Question Generation | CANARD Non-Biased | CQG Non-Biased Performance22.4 | 6 | |
| Conversational Question Generation | CANARD Biased | Performance26 | 6 | |
| Conversational Question Answering | CANARD Non-Biased | Performance (%)21.9 | 5 | |
| Conversational Question Answering | CANARD Biased | Performance67.5 | 5 | |
| Incomplete Utterance Restoration | CANARD (test) | BLEU59.92 | 5 | |
| Incomplete Utterance Restoration | CANARD (dev) | BLEU56.93 | 4 | |
| Incomplete Utterance Rewriting | CANARD | ROUGE-1 (B1)70.5 | 4 |
Showing 9 of 9 rows