Web Interaction on Mind2Web cross-domain
26.97Action AccuracyEvolving-RL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Evolving-RLSkill injection=true2026.05 | 26.97 | 1.91 | |
| Evolving-RLSkill injection=false2026.05 | 24.67 | 1.21 | |
| GRPOSkill injection=false2026.05 | 20.84 | 1.67 | |
| GRPOSkill injection=true2026.05 | 20.33 | 1.69 | |
| Memento2026.05 | 13.13 | 0.88 | |
| Base ModelSkill injection=true2026.05 | 12.68 | 0.5 | |
| ReasoningBank2026.05 | 12.09 | 0.66 | |
| Base ModelSkill injection=false2026.05 | 7.91 | 0.44 |