Ubuntu
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Response Selection | Ubuntu (test) | Recall@1 (Top 10)0.886 | 58 | |
| Response Selection | Ubuntu v2 (test) | MRR91.9 | 20 | |
| Dialogue Response Selection | Ubuntu (test) | R@1 (R10)0.867 | 18 | |
| Multi-party dialogue generation | Ubuntu IRC-16 | BLEU-116.04 | 12 | |
| Answer Ranking | Ubuntu v2 (test) | Recall@1 (1/2 Pool)91.5 | 11 | |
| Dialogue Retrieval | Ubuntu v2 | R@186.3 | 9 | |
| Conversation Disentanglement | Ubuntu (test) | VI91.5 | 8 | |
| Non-membership detection | Ubuntu | P-Value0.0001 | 5 | |
| Response Selection (Subtask 1) | Ubuntu (dev) | Recall@1 (R100)0.534 | 2 | |
| Dialogue Thread Reconstruction (Subtask 4) | Ubuntu Track 2 DSTC 8 (test) | Precision44.3 | 1 | |
| Response Selection with No Correct Response (Subtask 2) | Ubuntu Track 2 DSTC 8 (test) | Recall@150.6 | 1 | |
| Response Selection (Subtask 1) | Ubuntu Track 2 DSTC 8 (test) | Recall@164.9 | 1 | |
| Response Selection (Subtask 1) | Ubuntu (test) | Recall@1 (R100)60.8 | 1 |
Showing 13 of 13 rows