Deep Research on 2WIKI (test)
0.92Mean Correct RateOur base agent
Evaluation Results
| Method | Links | |
|---|---|---|
| Our base agentType=Live web2025.10 | 0.92 | |
| Our best agentType=Live web2025.10 | 0.908 | |
| WebSailorType=Live web2025.10 | 0.888 | |
| DeepResearcherType=Live web2025.10 | 0.854 | |
| ASearcherType=Live web2025.10 | 0.844 | |
| Search-R1Type=Static & simulated web2025.10 | 0.784 | |
| R1-SearcherType=Static & simulated web2025.10 | 0.616 | |
| ZeroSearchType=Static & simulated web2025.10 | 0.176 |