OpenDialKG
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Knowledge-intensive Dialogue | OpendialKG | Factual Accuracy88.84 | 11 | |
| Knowledge-grounded dialogue generation | OpenDialKG (test) | BLEU-420.77 | 10 | |
| Hallucination detection | OpenDialKG Eval (test) | Macro F176.2 | 7 | |
| Conversation | OpenDialKG | Dist-22.9162 | 7 | |
| Recommendation | OpenDialKG | Recall@128.95 | 7 | |
| RDF-to-text generation | OpenDialKG (test) | Grammaticality98.5 | 6 | |
| Knowledge-grounded Dialogue Generation | OpenDialKG | Faithfulness81.67 | 4 | |
| Knowledge-Grounded Dialogue Generation (Fluency) | OpenDialKG | Win Rate37.33 | 4 |
Showing 8 of 8 rows