Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Web Agent Navigation on Mind2Web Cross-Task 1.0

49.2Success Rate

PolySkill (+Update)

-1.96811.31624.637.884Oct 17, 2025Oct 25, 2025Nov 3, 2025Nov 12, 2025Nov 21, 2025Nov 30, 2025Dec 9, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
49.2--
2025.10
47.5--
2025.10
47--
2025.10
46.1--
2025.10
45.2--
2025.10
44.9--
2025.10
44.8--
2025.10
42.3--
2025.10
41.5--
2025.10
40.8--
2025.12
5.2854.945.1
2025.12
5.2153.245.2
2025.12
2.3844.237.6
2025.12
1.9848.234.3
2025.12
1.9834.831.1
2025.12
1.193126.1
2025.12
1.193522
2025.12
0.423.417.3
2025.12
07.26
2025.12
012.39.6
2025.12
011.49.1
2025.12
015.112.6
2025.12
07.13.5
2025.12
022.79.8
2025.12
017.413.6
2025.12
015.212