Student Simulation on Malrule (test)
96FT MetricQwen3-80B-Thinking
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen3-80B-ThinkingPrompting Strategy=Base Prompt2026.05 | 96 | 95 | 95 | 0.01 | |
| GPT-OSS-120BPrompting Strategy=Base Prompt2026.05 | 96 | 94 | 93 | 0.02 | |
| Qwen3-80B-ThinkingPrompting Strategy=Reflective Prompt2026.05 | 96 | 95 | 94 | 0.01 | |
| GPT-OSS-120BPrompting Strategy=Reflective Prompt2026.05 | 96 | 95 | 94 | 0.02 | |
| Qwen3-80B-InstructPrompting Strategy=Base Prompt2026.05 | 95 | 94 | 93 | 0.01 | |
| GPT-OSS-20BPrompting Strategy=Base Prompt2026.05 | 95 | 94 | 92 | 0.02 | |
| GPT-OSS-20BPrompting Strategy=Reflective Prompt2026.05 | 95 | 94 | 92 | 0.02 | |
| Qwen3-4BPrompting Strategy=Base Prompt2026.05 | 94 | 93 | 92 | 0.01 | |
| Qwen3-4BPrompting Strategy=Reflective Prompt2026.05 | 94 | 92 | 93 | 0.01 | |
| Qwen3-80B-InstructPrompting Strategy=Reflective Prompt2026.05 | 94 | 94 | 92 | 0.01 | |
| Llama3.3-70B-InstructPrompting Strategy=Base Prompt2026.05 | 91 | 89 | 73 | 0.1 | |
| Llama3.3-70B-InstructPrompting Strategy=Reflective Prompt2026.05 | 91 | 89 | 66 | 0.14 | |
| Llama3.1-8B-InstructPrompting Strategy=Base Prompt2026.05 | 73 | 68 | 59 | 0.09 | |
| Llama3.1-8B-InstructPrompting Strategy=Reflective Prompt2026.05 | 72 | 66 | 50 | 0.14 |