REDIAL
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Recommendation | REDIAL (test) | Recall@1050 | 46 | |
| Conversational Recommendation | REDIAL | Recall@50.2691 | 40 | |
| Conversational Performance | REDIAL (test) | Distinct-3122.7 | 37 | |
| Recommendation | ReDial | R@1022.4 | 24 | |
| Conversation | ReDial (test) | Fluency1.72 | 18 | |
| Response Generation | ReDial | Distinct-31.43 | 17 | |
| Conversational Recommendation | ReDial | User Satisfaction2.3 | 15 | |
| Sequential Recommendation | ReDial | NDCG@50.018 | 14 | |
| Conversation Performance | ReDial | BLEU-25.1 | 12 | |
| Recommendation | ReDial | Recall@10.083 | 12 | |
| Conversational Recommendation | ReDial (test) | Recall@15.6 | 9 | |
| User Satisfaction Estimation | ReDial 5% training size (test) | Precision71.2 | 8 | |
| Conversational Response Generation | ReDial (test) | Fluency2.6 | 7 | |
| Conversation | ReDial | Distinct-20.492 | 7 | |
| Recommendation | ReDial Cold start | R@13.9 | 7 | |
| Fair Recommendation | REDIAL (test) | A Score @50.0063 | 7 | |
| Knowledge Prediction | ReDial v1 (test) | Precision50 | 6 | |
| Response Generation | ReDial v1 (test) | BLEU28 | 6 | |
| Conversational Response Generation | ReDial | Fluency82 | 6 | |
| User Satisfaction Estimation | ReDial | Accuracy0.66 | 6 | |
| Response Generation | ReDial (Human Evaluation) | Relevance Score2.43 | 5 | |
| Conversational Recommendation | REDIAL | Recall@13 | 4 | |
| Conversational Recommendation | ReDial 1.0 (Offline) | R@14.46 | 4 | |
| Data2Text generation | ReDial (test) | Recall82 | 1 |
Showing 24 of 24 rows