Sentence Simplification (Lexical-Paraphrasing) on SimPA (out-of-domain)
37.3SARIPO_Think
Evaluation Results
| Method | Links | |
|---|---|---|
| PO_ThinkBackbone=Phi3.8B2025.12 | 37.3 | |
| PO_ThinkBackbone=Qwen14B2025.12 | 36.8 | |
| PO_ThinkBackbone=Llama8B2025.12 | 36.6 | |
| PO_ThinkBackbone=Qwen7B2025.12 | 36.4 | |
| GPT-4o2025.12 | 28.6 | |
| VanillaBackbone=Llama8B2025.12 | 27.7 | |
| VanillaBackbone=Qwen14B2025.12 | 27.5 | |
| VanillaBackbone=Qwen7B2025.12 | 26.5 | |
| VanillaBackbone=Phi3.8B2025.12 | 22.3 |