| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| NO-ConvAI2 NLEBench (test) | NorGPT-23B | BLEU4.28 | 7 | 3mo ago | |
| Open-domain dialogue Human-bot chat | PLATO-XL (Diamante) | Coherence1.92 | 5 | 3mo ago | |
| Chinese open-domain conversation Self-chat (test) | PLATO-XL (Diamante) | Coherence194.8 | 4 | 3mo ago |