General Evaluation on UltraFeedback Aggregate
59.59Overall Average ScoreLP-SFT
Evaluation Results
| Method | Links | |
|---|---|---|
| LP-SFTBackbone=Qwen3-14B2026.07 | 59.59 | |
| ASFTBackbone=Qwen3-14B2026.07 | 57.43 | |
| EAFTBackbone=Qwen3-14B2026.07 | 55.94 | |
| CEBackbone=Qwen3-14B2026.07 | 55.54 | |
| LP-SFTBackbone=Qwen3-4B2026.07 | 54.74 | |
| GEMBackbone=Qwen3-14B2026.07 | 53.55 | |
| EAFTBackbone=Qwen3-4B2026.07 | 51 | |
| ASFTBackbone=Qwen3-4B2026.07 | 50.53 | |
| CEBackbone=Qwen3-4B2026.07 | 49.9 | |
| GEMBackbone=Qwen3-4B2026.07 | 49.39 | |
| DFTBackbone=Qwen3-14B2026.07 | 48.47 | |
| DFTBackbone=Qwen3-4B2026.07 | 44.02 | |
| LP-SFTBackbone=Llama3.1-8B2026.07 | 34.01 | |
| GEMBackbone=Llama3.1-8B2026.07 | 32.86 | |
| CEBackbone=Llama3.1-8B2026.07 | 32.74 | |
| EAFTBackbone=Llama3.1-8B2026.07 | 31.79 | |
| ASFTBackbone=Llama3.1-8B2026.07 | 30.71 | |
| DFTBackbone=Llama3.1-8B2026.07 | 21.86 |