NLG Meta-evaluation on CUS-QA orig. (cs)
0.804Kendall CorrelationQwen 3 30B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen 3 30BShot=Zero2026.03 | 0.804 | |
| Llama 3.3 70BShot=Few2026.03 | 0.757 | |
| Llama 4 ScoutShot=Few2026.03 | 0.751 | |
| Qwen 3 30BShot=Few2026.03 | 0.693 | |
| Llama 4 ScoutShot=Zero2026.03 | 0.688 | |
| Llama 3.3 70BShot=Zero2026.03 | 0.672 |