Instruction Following Evaluation on Dolly Out-of-Distribution
49.9GPT-4o ScoreSODA
Evaluation Results
| Method | Links | |
|---|---|---|
| SODAModel=Llama-3.1-8B-Instruct2026.04 | 49.9 | |
| TeacherModel=GPT-5-Chat2026.04 | 49.8 | |
| SODAModel=Qwen2.5-7B-Instruct2026.04 | 49.6 | |
| SODAModel=Llama-3.2-3B-Instruct2026.04 | 49.5 | |
| GADModel=Llama-3.1-8B-Instruct2026.04 | 48.8 | |
| GADModel=Qwen2.5-7B-Instruct2026.04 | 48.5 | |
| GADModel=Llama-3.2-3B-Instruct2026.04 | 48.5 | |
| SeqKDModel=Llama-3.1-8B-Instruct2026.04 | 47.7 | |
| BaseModel=Qwen2.5-7B-Instruct2026.04 | 47.6 | |
| SeqKDModel=Qwen2.5-7B-Instruct2026.04 | 47.2 | |
| SeqKDModel=Llama-3.2-3B-Instruct2026.04 | 47 | |
| GADModel=Qwen2.5-3B-Instruct2026.04 | 46.7 | |
| BaseModel=Llama-3.1-8B-Instruct2026.04 | 46.6 | |
| SODAModel=Qwen2.5-3B-Instruct2026.04 | 46.1 | |
| BaseModel=Llama-3.2-3B-Instruct2026.04 | 45.8 | |
| BaseModel=Qwen2.5-3B-Instruct2026.04 | 45.1 | |
| SeqKDModel=Qwen2.5-3B-Instruct2026.04 | 44.8 |