Verifiable Instruction Following on IFEval (test)
75.23Prompt Loose AccuracyLLAMA-3.1-TULU-3-8B-DPO
Evaluation results
| Method | Links | |
|---|---|---|
| LLAMA-3.1-TULU-3-8B-DPOBase Model=Llama-3.1-8B, Fine-tuning=DPO2025.12 | 75.23 | |
| QWEN2.5-7B-INSTRUCTModel Size=7B, Type=Instruct2025.12 | 73.01 | |
| LLAMA-3.1-8B-INSTRUCTModel Size=8B, Type=Instruct2025.12 | 71.72 | |
| STACKELBERGGDA-LEADERRole=Leader2025.12 | 71.71 | |
| GEMMA-2-9B-ITModel Size=9B, Type=Instruct2025.12 | 71.53 | |
| LLAMA-3.1-TULU-3-8B-SFTBase Model=Llama-3.1-8B, Fine-tuning=SFT2025.12 | 67.46 | |
| STACKELBERGGDA-FOLLOWERRole=Follower2025.12 | 61.92 |