Summarization on XSUM (Win Rate)
80Win RateMultiMetric
Evaluation Results
| Method | Links | |
|---|---|---|
| MultiMetricModel=GPT-J, Baseline Comparison=MPO2026.05 | 80 | |
| MultiMetricModel=LLaMA, Baseline Comparison=MPO2026.05 | 54 | |
| MultiMetricModel=DeepSeek, Baseline Comparison=MPO2026.05 | 52.4 | |
| MultiMetricModel=BART, Baseline Comparison=MPO2026.05 | 52 | |
| MultiMetricModel=BART, Baseline Comparison=SFT2026.05 | 51.4 | |
| MultiMetricModel=GPT-J, Baseline Comparison=SFT2026.05 | 44.2 | |
| MultiMetricModel=LLaMA, Baseline Comparison=SFT2026.05 | 42 | |
| MultiMetricModel=DeepSeek, Baseline Comparison=SFT2026.05 | 39 |