Logical Fallacy Tutoring on Elec2Deb20 (normal students)
69.3DivergenceBASE
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| BASEBackbone=GPT-4o, Evaluator=QwQ-32B2026.05 | 69.3 | 8.7 | 13 | 54.9 | 11.1 | 43.4 | 49.2 | 4.4 | 31.2 | |
| BASE W/ PROBLEMSBackbone=GPT-4o, Evaluator=QwQ-32B2026.05 | 74 | 48.1 | 42.5 | 99.9 | 95.7 | 63 | 27.6 | 41.5 | 61.5 | |
| LFTutorBackbone=GPT-4o, Evaluator=QwQ-32B2026.05 | 84.6 | 87.9 | 78.3 | 99.6 | 96.1 | 91.2 | 95 | 43.6 | 84.5 |