Correctness simulation accuracy on DBE-KT22 (100% historical interaction logs)
73.03AccuracyEdu-Theater
Evaluation Results
| Method | Links | |
|---|---|---|
| Edu-TheaterBase LLM=GPT-4.1-mini2026.06 | 73.03 | |
| Agent4EduBase LLM=GPT-4.1-mini2026.06 | 71.3 | |
| EduAgentBase LLM=GPT-4.1-mini2026.06 | 69.95 | |
| Edu-TheaterBase LLM=Llama3-8b2026.06 | 69.77 | |
| Agent4EduBase LLM=Llama3-8b2026.06 | 69.5 | |
| EERNN2026.06 | 69.32 | |
| DAISIM2026.06 | 69.03 | |
| SAKT2026.06 | 68.92 | |
| EduAgentBase LLM=Llama3-8b2026.06 | 68.45 | |
| DKVMN2026.06 | 67.79 | |
| KES2026.06 | 65.41 |