Context-faithful Question Answering on CoConflictQA
82.43ConRPromptCD
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| PromptCDModel=LLAMA3-8B-INSTRUCT2026.02 | 82.43 | 15.76 | 16.05 | |
| PromptCDModel=QWEN2.5-7B-INSTRUCT2026.02 | 77.64 | 18.15 | 18.95 | |
| AttrModel=LLAMA3-8B-INSTRUCT2026.02 | 76.24 | 22.35 | 22.67 | |
| OpinModel=QWEN2.5-7B-INSTRUCT2026.02 | 76.24 | 21.75 | 22.2 | |
| VanillaModel=QWEN2.5-7B-INSTRUCT2026.02 | 75.84 | 24.15 | 24.15 | |
| AttrModel=QWEN2.5-7B-INSTRUCT2026.02 | 75.44 | 20.75 | 21.57 | |
| OpinModel=LLAMA3-8B-INSTRUCT2026.02 | 73.85 | 18.56 | 20.08 | |
| PromptCDModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 72.65 | 13.77 | 15.93 | |
| OpinModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 70.45 | 17.16 | 19.59 | |
| PromptCDModel=LLAMA2-7B-CHAT2026.02 | 70.25 | 11.16 | 13.71 | |
| VanillaModel=LLAMA3-8B-INSTRUCT2026.02 | 68.46 | 29.74 | 30.29 | |
| OpinModel=LLAMA2-7B-CHAT2026.02 | 67.26 | 13.57 | 16.79 | |
| AttrModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 67.07 | 16.37 | 19.62 | |
| VanillaModel=MISTRALV0.3-7B-INSTRUCT2026.02 | 67.06 | 19.96 | 22.94 | |
| AttrModel=LLAMA2-7B-CHAT2026.02 | 65.46 | 13.97 | 17.58 | |
| VanillaModel=LLAMA2-7B-CHAT2026.02 | 62.07 | 18.76 | 23.2 |