Multi-Hop Question Answering on Multi-Hop QA Full Context
86.06AccuracyQwen3-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-32BParameters=32B2026.01 | 86.06 | |
| GPT-4o2026.01 | 85.24 | |
| HyperClovaX-14BParameters=14B2026.01 | 84.47 | |
| Exaone3.5-7.8BParameters=7.8B2026.01 | 81.72 | |
| Qwen3-8BParameters=8B2026.01 | 79.46 |