Web interaction on Mind2Web (overall)
30.87Action AccuracyEvolving-RL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Evolving-RLSkill injection=true2026.05 | 30.87 | 1.99 | |
| Evolving-RLSkill injection=false2026.05 | 28.05 | 1.57 | |
| GRPOSkill injection=false2026.05 | 22.83 | 1.42 | |
| GRPOSkill injection=true2026.05 | 22.73 | 1.43 | |
| Memento2026.05 | 13.84 | 0.75 | |
| Base ModelSkill injection=true2026.05 | 13.41 | 0.42 | |
| ReasoningBank2026.05 | 12.79 | 0.52 | |
| Base ModelSkill injection=false2026.05 | 8.79 | 0.34 |