| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Upvote Prediction | Stackexchange | ROC-AUC85.02 | 45 | |
| Churn Prediction | Stackexchange | ROC-AUC84.22 | 45 | |
| Clustering | MTEB StackExchange P2P | V1 Score37.27 | 17 | |
| Clustering | MTEB StackExchange S2S | V1 Score61.49 | 17 | |
| Question Answering | StackExchange (test) | Accuracy65.6 | 12 | |
| user-churn | StackExchange 4DBInfer (test) | AUC0.8796 | 9 | |
| post-upvote | StackExchange 4DBInfer (test) | AUC0.8896 | 9 | |
| Question Answering | StackExchange Q&A (test) | Accuracy (Bio.)82.2 | 8 | |
| Present keyphrase generation | StackExchange | F1@327.2 | 8 | |
| Topic Classification | StackExchange (test) | Acc67.56 | 6 | |
| Language Modeling | StackExchange (val) | Perplexity4.43 | 3 | |
| Absent keyphrase generation | StackExchange | Recall@54.6 | 3 | |
| Reward Modeling | StackExchange (val) | REWARD0 | 1 |