Error Detection and Correction on KMP-Skills
86.3MR-ScoreGemini 2.0 Flash
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemini 2.0 FlashSize=-2026.03 | 86.3 | 87.5 | 96.7 | |
| LearnLM-1.5-ProSize=-2026.03 | 69.8 | 74.6 | 94.4 | |
| Qwen2.5-72B-InstSize=72B2026.03 | 65.8 | 78.6 | 89.5 | |
| GPT-4oSize=-2026.03 | 64.4 | 78.3 | 90.9 | |
| Qwen2.5-32B-InstSize=32B2026.03 | 64 | 69.8 | 87.8 | |
| Phi-4Size=14B2026.03 | 61.4 | 80.3 | 92.1 | |
| Qwen2.5-14B-InstSize=14B2026.03 | 58.2 | 70.9 | 82 | |
| GPT-4o miniSize=-2026.03 | 56.9 | 66.1 | 84.3 | |
| Qwen2.5-Math-72B-InstSize=72B2026.03 | 53.8 | 73.5 | 81 | |
| KMP-LM-7BSize=7B2026.03 | 49 | 57.5 | 63.4 | |
| Qwen2.5-7B-InstSize=7B2026.03 | 34.9 | 44.3 | 73.2 | |
| Llama-3.1-8B-InstSize=8B2026.03 | 28.6 | 9 | 55.9 | |
| TutorChat-LLMSize=7B2026.03 | 24 | 38 | 48.5 | |
| MathChatsync-LLMSize=7B2026.03 | 23.5 | 9.8 | 18.9 | |
| Mistral-Nemo-InstSize=12B2026.03 | 19.4 | 22.5 | 48.1 | |
| Qwen2.5-Math-7B-InstSize=7B2026.03 | 3.4 | 44.1 | 52.3 |