NLG Evaluation on SumPubMed
0.3Spearman CorrelationGPT-4o
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4oPrompting=Few-shot2026.02 | 0.3 | |
| MixtralPrompting=Zero-shot2026.02 | 0.3 | |
| MixtralPrompting=Few-shot2026.02 | 0.28 | |
| QwenPrompting=Few-shot2026.02 | 0.28 | |
| LlamaPrompting=Zero-shot2026.02 | 0.24 | |
| LlamaPrompting=Few-shot2026.02 | 0.2 | |
| QwenPrompting=Zero-shot2026.02 | 0.19 | |
| GPT-4oPrompting=Zero-shot2026.02 | 0.14 | |
| GPT-4o-miniPrompting=Few-shot2026.02 | 0.06 | |
| GPT-4o-miniPrompting=Zero-shot2026.02 | -0.09 |