Question Answering on GrailQA
77.03Exact Match (EM)Qwen3-235B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-235BFramework=Readi2026.04 | 77.03 | 31.13 | |
| MixedFramework=Readi2026.04 | 76.34 | 11 | |
| Qwen2.5-7BFramework=Readi2026.04 | 69.18 | — | |
| Qwen3-235BFramework=StructGPT2026.04 | 68.17 | 23.57 | |
| Qwen3-235BFramework=ToG2026.04 | 67.91 | 30.86 | |
| MixedFramework=StructGPT2026.04 | 67.79 | 10.04 | |
| MixedFramework=ToG2026.04 | 66.67 | 14.11 | |
| Qwen2.5-7BFramework=StructGPT2026.04 | 61.02 | — | |
| Qwen2.5-7BFramework=ToG2026.04 | 60.47 | — |