Personalized distractor generation on Discrete_40 Student_1361
40.4AccuracyMCTS-guided reasoning reconstruction framework
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MCTS-guided reasoning reconstruction frameworkBackbone=Claude-4-Sonnet2025.08 | 40.4 | 3.32 | 2.76 | |
| MCTS-guided reasoning reconstruction frameworkBackbone=Deepseek-V32025.08 | 34.7 | 3.24 | 2.69 | |
| MCTS-guided reasoning reconstruction frameworkBackbone=GPT-4o2025.08 | 31 | 3.05 | 2.52 | |
| MCTS-guided reasoning reconstruction frameworkBackbone=LLaMA-3-70B2025.08 | 28.6 | 2.82 | 2.33 | |
| MCTS-guided reasoning reconstruction frameworkBackbone=GPT-3.5-turbo2025.08 | 28.6 | 2.97 | 2.38 | |
| CoTBackbone=Claude-4-Sonnet2025.08 | 27.8 | 2.99 | 2.42 | |
| MasteryBackbone=Claude-4-Sonnet2025.08 | 26.5 | 2.9 | 2.31 | |
| IOBackbone=Claude-4-Sonnet2025.08 | 26.5 | 3.04 | 2.47 | |
| MasteryBackbone=Deepseek-V32025.08 | 22.9 | 2.82 | 2.27 | |
| IOBackbone=Deepseek-V32025.08 | 22 | 2.95 | 2.33 | |
| CoTBackbone=Deepseek-V32025.08 | 21.2 | 2.96 | 2.38 | |
| MasteryBackbone=GPT-3.5-turbo2025.08 | 18.8 | 2.6 | 2.07 | |
| IOBackbone=LLaMA-3-70B2025.08 | 18.4 | 2.54 | 2.11 | |
| Level+SimilarityBackbone=Claude-4-Sonnet2025.08 | 18.4 | 2.79 | 2.34 | |
| IOBackbone=GPT-3.5-turbo2025.08 | 18 | 2.6 | 2.1 | |
| CoTBackbone=GPT-3.5-turbo2025.08 | 18 | 2.64 | 2.16 | |
| CoTBackbone=GPT-4o2025.08 | 18 | 2.72 | 2.26 | |
| MasteryBackbone=LLaMA-3-70B2025.08 | 17.1 | 2.55 | 2.05 | |
| MasteryBackbone=GPT-4o2025.08 | 17.1 | 2.76 | 2.31 | |
| CoTBackbone=LLaMA-3-70B2025.08 | 16.7 | 2.59 | 2.15 | |
| SimilarityBackbone=Deepseek-V32025.08 | 16.7 | 2.62 | 2.12 | |
| IOBackbone=GPT-4o2025.08 | 16.3 | 2.69 | 2.26 | |
| Level+RandomBackbone=Deepseek-V32025.08 | 15.5 | 2.58 | 2.18 | |
| RandomBackbone=GPT-3.5-turbo2025.08 | 15.5 | 2.45 | 2.04 | |
| Level+RandomBackbone=GPT-3.5-turbo2025.08 | 15.5 | 2.36 | 1.98 | |
| LevelBackbone=GPT-4o2025.08 | 15.5 | 2.48 | 2.09 | |
| Level+RandomBackbone=LLaMA-3-70B2025.08 | 15.1 | 2.35 | 1.96 | |
| SimilarityBackbone=Claude-4-Sonnet2025.08 | 15.1 | 2.91 | 2.39 | |
| SimilarityBackbone=GPT-3.5-turbo2025.08 | 15.1 | 2.54 | 2.07 | |
| Level+RandomBackbone=GPT-4o2025.08 | 15.1 | 2.51 | 2.16 | |
| LevelBackbone=LLaMA-3-70B2025.08 | 14.7 | 2.34 | 1.93 | |
| LevelBackbone=Claude-4-Sonnet2025.08 | 14.7 | 2.75 | 2.38 | |
| Level+RandomBackbone=Claude-4-Sonnet2025.08 | 14.7 | 2.67 | 2.25 | |
| RandomBackbone=GPT-4o2025.08 | 14.3 | 2.49 | 2.08 | |
| LevelBackbone=GPT-3.5-turbo2025.08 | 13.9 | 2.36 | 1.9 | |
| Level+SimilarityBackbone=GPT-3.5-turbo2025.08 | 13.9 | 2.29 | 1.89 | |
| Level+SimilarityBackbone=GPT-4o2025.08 | 13.9 | 2.44 | 2.11 | |
| RandomBackbone=Deepseek-V32025.08 | 13.5 | 2.54 | 2.07 | |
| RandomBackbone=Claude-4-Sonnet2025.08 | 13.5 | 2.91 | 2.48 | |
| Level+SimilarityBackbone=LLaMA-3-70B2025.08 | 13.1 | 2.3 | 1.88 | |
| SimilarityBackbone=GPT-4o2025.08 | 12.7 | 2.63 | 2.24 | |
| SimilarityBackbone=LLaMA-3-70B2025.08 | 11.4 | 2.3 | 1.92 | |
| LevelBackbone=Deepseek-V32025.08 | 11 | 2.49 | 2.08 | |
| Level+SimilarityBackbone=Deepseek-V32025.08 | 10.6 | 2.44 | 1.99 | |
| RandomBackbone=LLaMA-3-70B2025.08 | 9 | 2.17 | 1.78 |