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

Language Agent Task on ScienceWorld (OOD)

29.6Score

Task-Perturbed NLL Optimization

24.9226.13527.3528.565Jun 25, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
29.6
2026.06
26.4
2026.06
25.6
2026.06
25.1