Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Web interaction on Mind2Web (cross-website)
Loading...
35.15
Action Accuracy
Evolving-RL
6.862
14.206
21.55
28.894
May 11, 2026
Action Accuracy
Success Rate
Updated 21d ago
Evaluation Results
Method
Method
Links
Action Accuracy
Success Rate
Evolving-RL
Skill injection=false
2026.05
35.15
3.39
Evolving-RL
Skill injection=true
2026.05
35.14
1.81
GRPO
Skill injection=true
2026.05
26.2
0.9
GRPO
Skill injection=false
2026.05
24.61
0
ReasoningBank
2026.05
17.76
0.56
Memento
2026.05
15.46
0.56
Base Model
Skill injection=true
2026.05
15.01
0
Base Model
Skill injection=false
2026.05
7.95
0
Feedback
Search any
task
Search any
task