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

Agentic Task Completion on ScienceWorld

75L0 Score

SPARK

-1.75218.17438.158.026Jan 28, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
7574.157.8
2026.01
69.56849.2
2026.01
67.24332.2
2026.01
62.540.628.1
2026.01
53.933.633.6
2026.01
53.435.225.8
2026.01
49.130.126.6
2026.01
46.934.426.5
2026.01
39.831.331.4
2026.01
39.122.715.6
2026.01
36.73223.4
2026.01
32.827.330.5
2026.01
3232.832.8
2026.01
25.815.24.7
2026.01
21.113.710.9
2026.01
20.31812.5
2026.01
7.811.36.3
2026.01
1.20.80.8