Instruction Following on UltraFeedback
80.8Win RateTeacher
Evaluation Results
| Method | Links | |
|---|---|---|
| TeacherModel=Qwen2.5-7B-Instruct, Role=Teacher2025.10 | 80.8 | |
| DistiLLM-2alpha=-5, Base Model=Qwen2.5-1.5B-Instruct2025.10 | 63.3 | |
| DistiLLM-2alpha=-1, Base Model=Qwen2.5-1.5B-Instruct2025.10 | 62.6 | |
| StudentModel=Qwen2.5-1.5B-Instruct, Role=Student2025.10 | 40 |