Pedagogical Response Evaluation on KMP-Dialogue
87.1Overall Judgment ScoreClaude-3.7-Sonnet
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Claude-3.7-SonnetAccess=Closed-Source Models2026.03 | 87.1 | 69.8 | 80 | 79.3 | 71.2 | 69.1 | 79.3 | 72.2 | 72.5 | |
| Claude-3.7-SonnetAccess=Closed-Source2026.03 | 87.1 | 69.8 | 80 | 79.3 | 71.2 | 69.1 | 79.3 | 72.2 | 72.5 | |
| DeepSeek-V3Access=Open-Source Models2026.03 | 84.6 | 71.6 | 70.3 | 81.5 | 71.4 | 69.6 | 83.9 | 70.8 | 73.1 | |
| DeepSeek-V3Access=Open-Source2026.03 | 84.6 | 71.6 | 70.3 | 81.5 | 71.4 | 69.6 | 83.9 | 70.8 | 73.1 | |
| LLaMa-3.3-70B-InstAccess=Open-Source Models2026.03 | 77.2 | 63.7 | 59.2 | 67.5 | 48.8 | 60.4 | 81.2 | 70.4 | 64.2 | |
| LearnLM-2.0-FlashAccess=Closed-Source Models2026.03 | 74.8 | 58.9 | 77.3 | 42.7 | 24.9 | 67.6 | 73.5 | 63.9 | 58.6 | |
| LearnLM-2.0-FlashAccess=Closed-Source2026.03 | 74.8 | 58.9 | 77.3 | 42.7 | 24.9 | 67.6 | 73.5 | 63.9 | 58.6 | |
| Qwen2.5-72B-InstAccess=Open-Source Models2026.03 | 68.5 | 55.5 | 50.5 | 77.3 | 71.5 | 60.7 | 54 | 52.4 | 58.3 | |
| Qwen2.5-72B-InstAccess=Open-Source2026.03 | 68.5 | 55.5 | 50.5 | 77.3 | 71.5 | 60.7 | 54 | 52.4 | 58.3 | |
| Qwen2.5-32B-InsAccess=Open-Source Models2026.03 | 67.5 | 55 | 52 | 71.6 | 70.7 | 61.8 | 55.5 | 54.1 | 58 | |
| Mistral-Nemo-InstAccess=Open-Source Models2026.03 | 66.1 | 56.3 | 53.5 | 64 | 54.8 | 56.1 | 57.9 | 61.1 | 57.1 | |
| Qwen2.5-14B-InstAccess=Open-Source Models2026.03 | 65 | 52.8 | 48.7 | 72 | 68.2 | 54.7 | 55.9 | 51.9 | 55.7 | |
| Phi-4Access=Open-Source Models2026.03 | 64 | 52.7 | 41.3 | 73.1 | 70 | 53 | 54.4 | 51 | 54.9 | |
| Qwen2.5-7B-InstAccess=Open-Source Models2026.03 | 62.5 | 52.9 | 41.2 | 74.2 | 66.5 | 56.6 | 49.7 | 50.6 | 54.7 | |
| Qwen2.5-7B-InstAccess=Open-Source2026.03 | 62.5 | 52.9 | 41.2 | 74.2 | 66.5 | 56.6 | 49.7 | 50.6 | 54.7 | |
| GPT-4oAccess=Closed-Source Models2026.03 | 60.7 | 44.5 | 43.4 | 62.2 | 57.5 | 47.5 | 59.9 | 41.3 | 48.2 | |
| GPT-4oAccess=Closed-Source2026.03 | 60.7 | 44.5 | 43.4 | 62.2 | 57.5 | 47.5 | 59.9 | 41.3 | 48.2 | |
| Gemini-2.0-FlashAccess=Closed-Source Models2026.03 | 58.5 | 38 | 51.1 | 49.9 | 34.2 | 42.2 | 50.8 | 37.6 | 41.1 | |
| Gemini-2.0-FlashAccess=Closed-Source2026.03 | 58.5 | 38 | 51.1 | 49.9 | 34.2 | 42.2 | 50.8 | 37.6 | 41.1 | |
| LLaMa-3.1-8BAccess=Open-Source Models2026.03 | 51.3 | 40.3 | 33 | 50.1 | 29 | 38.3 | 55 | 46.2 | 41.1 | |
| LLaMa-3.1-8BAccess=Open-Source2026.03 | 51.3 | 40.3 | 33 | 50.1 | 29 | 38.3 | 55 | 46.2 | 41.1 | |
| KMP-LM-7BAccess=Open-Source Models2026.03 | 40.4 | 35.1 | 29.3 | 47 | 38.5 | 39.8 | 43 | 36.1 | 37 | |
| KMP-LM-7BAccess=Open-Source, Base Model=Qwen2.5-Math-7B2026.03 | 40.4 | 35.1 | 29.3 | 47 | 38.5 | 39.8 | 43 | 36.1 | 37 | |
| TutorChat-LLMAccess=Open-Source Models2026.03 | 32.1 | 30 | 21.1 | 41.3 | 33.2 | 37.1 | 40.6 | 29.9 | 31.9 | |
| TutorChat-LLMAccess=Open-Source, Base Model=Qwen2.5-Math-7B2026.03 | 32.1 | 30 | 21.1 | 41.3 | 33.2 | 37.1 | 40.6 | 29.9 | 31.9 | |
| Qwen2.5-Math-72B-InstAccess=Open-Source Models2026.03 | 31 | 27.1 | 25.6 | 51.3 | 42.9 | 33.9 | 29.5 | 23.7 | 30.8 | |
| Qwen2.5-Math-72B-InstAccess=Open-Source2026.03 | 31 | 27.1 | 25.6 | 51.3 | 42.9 | 33.9 | 29.5 | 23.7 | 30.8 | |
| SocraticLMAccess=Open-Source Models2026.03 | 21.2 | 15.4 | 17 | 25 | 21 | 21.4 | 27.3 | 17.8 | 18.5 | |
| SocraticLMAccess=Open-Source2026.03 | 21.2 | 15.4 | 17 | 25 | 21 | 21.4 | 27.3 | 17.8 | 18.5 | |
| Qwen2.5-Math-7B-InstAccess=Open-Source Models2026.03 | 21 | 19.7 | 17.5 | 41.7 | 34.5 | 23.2 | 25.5 | 22.6 | 23.6 | |
| Qwen2.5-Math-7B-InstAccess=Open-Source2026.03 | 21 | 19.7 | 17.5 | 41.7 | 34.5 | 23.2 | 25.5 | 22.6 | 23.6 | |
| MathChatsync-LLMAccess=Open-Source Models2026.03 | 16.1 | 12 | 16 | 23 | 15.4 | 17.6 | 23.1 | 16.3 | 15.3 | |
| MathChatsync-LLMAccess=Open-Source, Base Model=Qwen2.5-Math-7B2026.03 | 16.1 | 12 | 16 | 23 | 15.4 | 17.6 | 23.1 | 16.3 | 15.3 |