ScienceWorld
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
ScienceWorld
56.67Success Rate (Task 6-1)
5
ScienceWorld
53.44Mean Score
5
ScienceWorld
72.67Task 5-2 (L) Mean Score
5
ScienceWorld
50Mean Score (Task 5-1 L)
5
ScienceWorld
40.83Mean Score
5
ScienceWorld
100Mean Score (Task 4-4 S)
5
ScienceWorld
100Mean Score (Task 4-3 (S))
5
ScienceWorld
100Mean Score
5
ScienceWorld
100Mean Score
5
ScienceWorld
95.83Mean Score
5
ScienceWorld
58.33Mean Score
5
ScienceWorld
68.67Mean Performance
5
ScienceWorld
82.33Mean Score (Task 3-1 S)
5
ScienceWorld
68.5Mean Score
5
ScienceWorld
92.33Mean Score (Task 2-3 L)
5
ScienceWorld
79.67Mean Score (Task 2-2)
5
ScienceWorld
83Mean Score
5
ScienceWorld
85Mean Score
5
ScienceWorld
70.67Mean Score (Task 1-4 L)
5
ScienceWorld
32Mean Score (Task 1-3 L)
5
ScienceWorld
70Mean Score
5
ScienceWorld
27.67Mean Score (Task 1-1 L)
5
ScienceWorld
49.58Mean Score
5
ScienceWorld 20 unseen variants (test)
83.7Return
5
ScienceWorld 20 unseen variants (test)
97.5Return
5