Deep Research on TriviaQA (TQ) (test)
92.8Mean Correct Answer RateWebSailor
Evaluation results
| Method | Links | |
|---|---|---|
| WebSailorType=Live web2025.10 | 92.8 | |
| Our best agentType=Live web2025.10 | 92.6 | |
| ASearcherType=Live web2025.10 | 84.6 | |
| Our base agentType=Live web2025.10 | 82.7 | |
| DeepResearcherType=Live web2025.10 | 79.8 | |
| Search-R1Type=Static & simulated web2025.10 | 74.2 | |
| R1-SearcherType=Static & simulated web2025.10 | 65 | |
| ZeroSearchType=Static & simulated web2025.10 | 31.4 |