Question Answering on HLE (test)
16.6EMWebExpert+SFT
Evaluation Results
| Method | Links | |
|---|---|---|
| WebExpert+SFTBackbone=QwQ-32B, Training=SFT2026.02 | 16.6 | |
| WebExpertBackbone=QwQ-32B2026.02 | 14.5 | |
| WebThinker-32B-BaseBackbone=32B-Base2026.02 | 13 | |
| Search-o1-32BBackbone=o1-32B2026.02 | 10.8 | |
| RAG-QwQ-32BStrategy=RAG2026.02 | 7.2 | |
| QwQ-32BStrategy=Direct2026.02 | 5.4 |