Trace-quality Evaluation on Trace-quality Dataset (evaluation set)
7.96FactualityClaude S. 4.6
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Claude S. 4.62026.06 | 7.96 | 7.91 | 7.89 | 7.93 | 7.98 | 7.95 | 7.94 | 391 | |
| GPT-4o2026.06 | 7.8 | 7.39 | 7.78 | 7.07 | 7.75 | 7.63 | 7.57 | 392 | |
| Gemini 2.52026.06 | 7.69 | 7.21 | 7.15 | 7.08 | 7.57 | 7.64 | 7.39 | 399 | |
| MARD-7B2026.06 | 7.17 | 6.05 | 7.29 | 7.03 | 6.8 | 6.49 | 6.8 | 586 |