Instruction Following on FollowRAG Instruction v1 (test)
41.1FollowRAG Instruction ScoreNuggets
Evaluation Results
| Method | Links | |
|---|---|---|
| NuggetsBase Model=Qwen-2, Data Ratio=10%2025.02 | 41.1 | |
| CaRBase Model=Qwen-2, Data Ratio=10%2025.02 | 40.5 | |
| IFDBase Model=Qwen-2, Data Ratio=15%2025.02 | 40.5 | |
| NuggetsBase Model=Qwen-2, Data Ratio=15%2025.02 | 40.3 | |
| NuggetsBase Model=Qwen-2, Data Ratio=5%2025.02 | 40.2 | |
| IFDBase Model=Qwen-2, Data Ratio=10%2025.02 | 40.1 | |
| NOVABase Model=Qwen-2, Data Ratio=10%2025.02 | 40.1 | |
| NOVABase Model=Qwen-2, Data Ratio=15%2025.02 | 40 | |
| CaRBase Model=Qwen-2, Data Ratio=15%2025.02 | 39.8 | |
| NOVABase Model=Qwen-2, Data Ratio=5%2025.02 | 39.7 | |
| CaRBase Model=Qwen-2, Data Ratio=5%2025.02 | 39.5 | |
| IFDBase Model=Qwen-2, Data Ratio=5%2025.02 | 39.2 | |
| NOVABase Model=LLaMA-1, Data Ratio=10%2025.02 | 39 | |
| NuggetsBase Model=LLaMA-1, Data Ratio=10%2025.02 | 38.7 | |
| FLAME-DPOBase Model=Qwen-22025.02 | 38.7 | |
| IFDBase Model=LLaMA-1, Data Ratio=10%2025.02 | 38.6 | |
| NuggetsBase Model=LLaMA-1, Data Ratio=15%2025.02 | 38.6 | |
| CaRBase Model=LLaMA-1, Data Ratio=5%2025.02 | 38.2 | |
| IFDBase Model=LLaMA-1, Data Ratio=15%2025.02 | 38.2 | |
| VanillaBase Model=Qwen-2, Data Ratio=100%2025.02 | 38.2 | |
| SELF-EVALBase Model=LLaMA-12025.02 | 38.1 | |
| NOVABase Model=LLaMA-1, Data Ratio=5%2025.02 | 38.1 | |
| CaRBase Model=LLaMA-1, Data Ratio=10%2025.02 | 38.1 | |
| NuggetsBase Model=LLaMA-1, Data Ratio=5%2025.02 | 38 | |
| NOVABase Model=LLaMA-1, Data Ratio=15%2025.02 | 38 | |
| IFDBase Model=LLaMA-1, Data Ratio=5%2025.02 | 37.8 | |
| VanillaBase Model=LLaMA-1, Data Ratio=100%2025.02 | 37.7 | |
| CaRBase Model=LLaMA-1, Data Ratio=15%2025.02 | 37.6 | |
| FLAME-DPOBase Model=LLaMA-12025.02 | 37.5 | |
| SELF-EVALBase Model=Qwen-22025.02 | 37.3 |