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

Science Experiment Execution on ScienceWorld (test)

51.51Success Rate

G-Mem

12.832422.873732.91542.9563May 8, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2026.05
51.51
2026.05
50.82
2026.05
50
2026.05
48.44
2026.05
47.78
2026.05
46.67
2026.05
46.67
2026.05
45.55
2026.05
44.44
2026.05
42.86
2026.05
34.44
2026.05
32.22
2026.05
32.22
2026.05
32.22
2026.05
31.11
2026.05
27.78
2026.05
26.67
2026.05
26.67
2026.05
25.55
2026.05
24.65
2026.05
22.22
2026.05
21.11
2026.05
15.56
2026.05
14.32