Instruction Following on VicunaEval (Score Metric)
40.75VicunaEval ScoreIOA
Evaluation Results
| Method | Links | |
|---|---|---|
| IOAModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 40.75 | |
| IOAStudent Model=Qwen2.5-14B, Teacher LLM=DeepSeek-R12026.02 | 39.97 | |
| IOA (Ours)Model=Qwen2.5-7B2026.02 | 39.97 | |
| IOAModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 39.13 | |
| MADAModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 38.99 | |
| Star-AgentsModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 38.59 | |
| IOAStudent Model=Qwen2.5-7B, Teacher LLM=DeepSeek-R12026.02 | 38.39 | |
| CasCoDModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 37.81 | |
| POCLModel=Qwen2.5-7B2026.02 | 37.61 | |
| MADAModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 37.4 | |
| DSSModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 37.35 | |
| GKDModel=Qwen2.5-7B2026.02 | 37.2 | |
| Star-AgentsModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 37 | |
| CounterDistillModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 36.98 | |
| SuperCorrectModel=Qwen2.5-7B2026.02 | 36.98 | |
| IOAModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 36.83 | |
| DistiLLM-2Model=Qwen2.5-7B2026.02 | 36.8 | |
| LionModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 36.52 | |
| IOAModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 36.41 | |
| IOAStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 36.29 | |
| CasCoDModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 36.21 | |
| IOAStudent Model=LLaMA3.1-8B, Teacher LLM=DeepSeek-R12026.02 | 36.12 | |
| ABKDModel=Qwen2.5-7B2026.02 | 36.01 | |
| DSSModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 35.75 | |
| MADAStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 35.42 | |
| CounterDistillModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 35.38 | |
| MADAModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 35.09 | |
| Star-AgentsStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 35.07 | |
| LaMiniModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 35.05 | |
| LionModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 34.92 | |
| IOAModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 34.81 | |
| MADAModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 34.81 | |
| IOAStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 34.72 | |
| Star-AgentsModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 34.69 | |
| CasCoDStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 34.25 | |
| Star-AgentsModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 34.23 | |
| CasCoDModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 33.91 | |
| DSSStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 33.81 | |
| MADAStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 33.63 | |
| CasCoDModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 33.53 | |
| DSSModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 33.45 | |
| LaMiniModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 33.45 | |
| CounterDistillStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 33.37 | |
| Self-InstructModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 33.26 | |
| MADAModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 33.21 | |
| DSSModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 33.17 | |
| CounterDistillModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 33.08 | |
| Star-AgentsStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 33.01 | |
| LionStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 32.93 | |
| CounterDistillModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 32.72 | |
| Star-AgentsModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 32.63 | |
| LionModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 32.62 | |
| LionModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 32.46 | |
| CasCoDStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 32.28 | |
| CasCoDModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 31.93 | |
| DSSStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 31.93 | |
| Self-InstructModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 31.66 | |
| DSSModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 31.57 | |
| CounterDistillStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 31.47 | |
| LaMiniModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 31.25 | |
| LaMiniModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 31.15 | |
| LaMiniStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 31.13 | |
| CounterDistillModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 31.12 | |
| LionStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 31.02 | |
| LionModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 30.86 | |
| LaMiniModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 29.65 | |
| LaMiniStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 29.58 | |
| Self-InstructModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 29.44 | |
| Self-InstructModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 29.36 | |
| Self-InstructStudent Model=Qwen2.5-3B, Teacher Model=DeepSeek-R12026.02 | 29.35 | |
| Self-InstructModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 27.84 | |
| Self-InstructStudent Model=LLaMA3.2-3B, Teacher Model=DeepSeek-R12026.02 | 27.83 | |
| UndistilledModel=Qwen2.5-14B, Teacher Model=OpenAI o12026.02 | 27.72 | |
| UndistilledModel=Qwen2.5-7B, Teacher Model=OpenAI o12026.02 | 26.12 | |
| UndistilledStudent Model=Qwen2.5-7B2026.02 | 25.92 | |
| UndistilledStudent Model=Qwen2.5-3B, Teacher Model=None2026.02 | 23.84 | |
| UndistilledModel=Qwen2.5-3B, Teacher LLM=OpenAI o12026.02 | 23.82 | |
| UndistilledModel=LLaMA3.1-8B, Teacher Model=OpenAI o12026.02 | 23.57 | |
| UndistilledStudent Model=LLaMA3.2-3B, Teacher Model=None2026.02 | 22.01 | |
| UndistilledModel=LLaMA3.2-3B, Teacher LLM=OpenAI o12026.02 | 21.97 |