Multi-hop Question Answering on HotpotQA (val) (F1, EM, p)
23.56F1 ScoreQwen3-32B-DS-SFT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen3-32B-DS-SFTSetting=Zero-shot, Mode=Closed-book, Teacher LLM=DeepSeek2026.05 | 23.56 | 13.09 | 0.021 | |
| Qwen3-32B-GE-SFTSetting=Zero-shot, Mode=Closed-book, Teacher LLM=Gemini2026.05 | 23.53 | 12.97 | 0.035 | |
| Qwen3-32B-BaseSetting=Zero-shot, Mode=Closed-book2026.05 | 22.97 | 12.43 | — |