Advanced Question Answering on WebWalkerQA
47.4Exact MatchSynPlanResearch-R1
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SynPlanResearch-R1Model Backbone=Qwen3-8B2026.03 | 47.4 | — | |
| SynPlanResearch-R1Model Backbone=Qwen3-4B2026.03 | 43.8 | — | |
| SimpleDeepSearcherModel Backbone=Qwen3-8B2026.03 | 42.4 | — | |
| SimpleDeepSearcherModel Backbone=Qwen3-4B2026.03 | 40.3 | — | |
| Search-R1Model Backbone=Qwen3-8B2026.03 | 37.8 | — | |
| Search-R1Model Backbone=Qwen3-4B2026.03 | 35.8 | — | |
| Rejection SamplingModel Backbone=Qwen3-8B2026.03 | 35.3 | — | |
| Rejection SamplingModel Backbone=Qwen3-4B2026.03 | 34.5 | — | |
| Search-o1Model Backbone=Qwen3-8B2026.03 | 33.9 | — | |
| RAGModel Backbone=Qwen3-4B2026.03 | 32.6 | — | |
| Search-o1Model Backbone=Qwen3-4B2026.03 | 32.3 | — | |
| RAGModel Backbone=Qwen3-8B2026.03 | 28.7 | — | |
| Direct InferenceModel Backbone=Qwen3-4B2026.03 | 17.9 | — | |
| Direct InferenceModel Backbone=Qwen3-8B2026.03 | 14 | — |