Knowledge-Intensive Reasoning on 2Wiki
0.89Average ScoreAutoTraj
Evaluation Results
| Method | Links | |
|---|---|---|
| AutoTrajCategory=SFT-RL TIR Methods2026.01 | 0.89 | |
| R1-SearcherCategory=RL-only TIR Methods2026.01 | 0.87 | |
| AutoTIRCategory=RL-only TIR Methods2026.01 | 0.86 | |
| ReSearchCategory=RL-only TIR Methods2026.01 | 0.81 | |
| Tool-Star-SFTCategory=SFT-only TIR Methods2026.01 | 0.79 | |
| Qwen2.5-7B-InstructFramework=Multi-Dimensional Trajectory Evaluation2026.01 | 0.78 | |
| Tool-StarCategory=SFT-RL TIR Methods2026.01 | 0.77 | |
| Vanilla SFT-RL TIRCategory=SFT-RL TIR Methods2026.01 | 0.75 | |
| ToRLCategory=RL-only TIR Methods2026.01 | 0.73 |