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

Interactive Task Completion on ScienceWorld

67.2L0 Score

RLVMR

19.25631.70344.1556.597Apr 27, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
67.24332.247.5
2026.04
66.666.55161.4
2026.04
62.540.628.143.7
2026.04
59.858.134.250.7
2026.04
53.425.225.834.8
2026.04
49.130.126.635.3
2026.04
46.934.326.535.9
2026.04
45.449.24145.2
2026.04
39.122.715.625.8
2026.04
27.335.226.529.7
2026.04
25.815.24.715.2
2026.04
22.231.429.127.6
2026.04
21.113.710.915.2