Open-ended Generation on TruthfulQA
70.13BLEURT ScoreFineSteer
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FineSteerBase Model=Gemma-2-9B-IT2026.04 | 70.13 | 70.9 | |
| TruthFlowBase Model=Gemma-2-9B-IT2026.04 | 68.74 | 66.01 | |
| FineSteerBase Model=Llama-3-8B-Instruct2026.04 | 66.5 | 62.35 | |
| FineSteerBase Model=Llama-3-8B-Instruct2026.04 | 66.5 | 62.35 | |
| FineSteerBase Model=Llama-3.2-3B-Instruct2026.04 | 65.31 | 54.79 | |
| AlphaSteerBase Model=Gemma-2-9B-IT2026.04 | 64.53 | 61.86 | |
| BiPOBase Model=Gemma-2-9B-IT2026.04 | 64.15 | 62.59 | |
| ITIBase Model=Gemma-2-9B-IT2026.04 | 63.79 | 61.37 | |
| FineSteerBase Model=Qwen2.5-7B-Instruct2026.04 | 63.57 | 54.28 | |
| FineSteerBase Model=Qwen2.5-7B-Instruct2026.04 | 63.57 | 54.28 | |
| CASTBase Model=Gemma-2-9B-IT2026.04 | 62.77 | 60.23 | |
| TruthFlowBase Model=Llama-3.2-3B-Instruct2026.04 | 62.73 | 50.61 | |
| Gemma-2-9B-IT2026.04 | 62.34 | 58.92 | |
| TruthFlowBase Model=Qwen2.5-7B-Instruct2026.04 | 62.1 | 49.88 | |
| TruthFlowBase Model=Qwen2.5-7B-Instruct2026.04 | 62.1 | 49.88 | |
| TruthFlowBase Model=Llama-3-8B-Instruct2026.04 | 61.66 | 54.77 | |
| TruthFlowBase Model=Llama-3-8B-Instruct2026.04 | 61.66 | 54.77 | |
| DoLaBase Model=Gemma-2-9B-IT2026.04 | 61.53 | 59.9 | |
| ITIBase Model=Qwen2.5-7B-Instruct2026.04 | 60.17 | 47.92 | |
| ITIBase Model=Qwen2.5-7B-Instruct2026.04 | 60.17 | 47.92 | |
| CASTBase Model=Qwen2.5-7B-Instruct2026.04 | 59.66 | 48.17 | |
| CASTBase Model=Qwen2.5-7B-Instruct2026.04 | 59.66 | 48.17 | |
| Qwen2.5-7B-InstructBase Model=Qwen2.5-7B-Instruct2026.04 | 59.41 | 48.41 | |
| Qwen2.5-7B-Instruct2026.04 | 59.41 | 48.41 | |
| AlphaSteerBase Model=Qwen2.5-7B-Instruct2026.04 | 58.68 | 48.66 | |
| AlphaSteerBase Model=Qwen2.5-7B-Instruct2026.04 | 58.68 | 48.66 | |
| BiPOBase Model=Qwen2.5-7B-Instruct2026.04 | 58.44 | 47.92 | |
| BiPOBase Model=Qwen2.5-7B-Instruct2026.04 | 58.44 | 47.92 | |
| DoLaBase Model=Qwen2.5-7B-Instruct2026.04 | 57.21 | 47.43 | |
| AlphaSteerBase Model=Llama-3.2-3B-Instruct2026.04 | 57.21 | 50.37 | |
| DoLaBase Model=Qwen2.5-7B-Instruct2026.04 | 57.21 | 47.43 | |
| CASTBase Model=Llama-3.2-3B-Instruct2026.04 | 56.72 | 47.19 | |
| BiPOBase Model=Llama-3.2-3B-Instruct2026.04 | 55.74 | 47.95 | |
| CASTBase Model=Llama-3-8B-Instruct2026.04 | 54.92 | 50.32 | |
| CASTBase Model=Llama-3-8B-Instruct2026.04 | 54.92 | 50.32 | |
| DoLaBase Model=Llama-3.2-3B-Instruct2026.04 | 54.28 | 42.05 | |
| ITIBase Model=Llama-3.2-3B-Instruct2026.04 | 53.66 | 45.97 | |
| AlphaSteerBase Model=Llama-3-8B-Instruct2026.04 | 53.42 | 51.83 | |
| AlphaSteerBase Model=Llama-3-8B-Instruct2026.04 | 53.42 | 51.83 | |
| Llama-3.2-3B-Instruct2026.04 | 52.09 | 41.08 | |
| BiPOBase Model=Llama-3-8B-Instruct2026.04 | 51.98 | 48.9 | |
| BiPOBase Model=Llama-3-8B-Instruct2026.04 | 51.98 | 48.9 | |
| DoLaBase Model=Llama-3-8B-Instruct2026.04 | 51.34 | 47.43 | |
| DoLaBase Model=Llama-3-8B-Instruct2026.04 | 51.34 | 47.43 | |
| ITIBase Model=Llama-3-8B-Instruct2026.04 | 51.23 | 50.37 | |
| ITIBase Model=Llama-3-8B-Instruct2026.04 | 51.23 | 50.37 | |
| Llama-3-8B-InstructBase Model=Llama-3-8B-Instruct2026.04 | 51.1 | 45.48 | |
| Llama-3-8B-Instruct2026.04 | 51.1 | 45.48 |