Story Generation on Story Generation Gender (UK) Real distribution
0.21MAEZero
Evaluation Results
| Method | Links | |
|---|---|---|
| ZeroBackbone=Llama1B2026.04 | 0.21 | |
| OursBackbone=Llama1B2026.04 | 0.22 | |
| DPOBackbone=Qwen7B2026.04 | 0.23 | |
| ExpBackbone=Llama1B2026.04 | 0.24 | |
| OursBackbone=Llama8B2026.04 | 0.25 | |
| ZeroBackbone=Qwen7B2026.04 | 0.26 | |
| ExpBackbone=Qwen7B2026.04 | 0.26 | |
| DPOBackbone=Qwen1.5B2026.04 | 0.26 | |
| IFTBackbone=Llama1B2026.04 | 0.26 | |
| DPOBackbone=Llama1B2026.04 | 0.26 | |
| IFTBackbone=Qwen7B2026.04 | 0.27 | |
| OursBackbone=Qwen7B2026.04 | 0.27 | |
| ZeroBackbone=Qwen1.5B2026.04 | 0.27 | |
| IFTBackbone=Qwen1.5B2026.04 | 0.27 | |
| IFTBackbone=Llama8B2026.04 | 0.27 | |
| OursBackbone=Qwen1.5B2026.04 | 0.28 | |
| ZeroBackbone=Llama8B2026.04 | 0.28 | |
| DPOBackbone=Llama8B2026.04 | 0.28 | |
| ImpBackbone=Llama1B2026.04 | 0.31 | |
| ExpBackbone=Qwen1.5B2026.04 | 0.32 | |
| ExpBackbone=Llama8B2026.04 | 0.32 | |
| ImpBackbone=Qwen1.5B2026.04 | 0.4 | |
| ImpBackbone=Qwen7B2026.04 | 0.44 | |
| ImpBackbone=Llama8B2026.04 | 0.46 |