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

Multi-Armed Bandit (MAB) Horizon Generalization T=100

22.37Average Regret

Iterative RMFT

20.91630.730540.54550.3595Nov 6, 2025
Updated 2d ago

Evaluation Results

MethodLinks
2025.11
22.37
2025.11
32.97
2025.11
33.44
2025.11
35.88
2025.11
40.09
2025.11
52.1
2025.11
58.72