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

Multi-fidelity Multi-armed Bandit on NLI shared evaluation pool

3,277.7Mean Cost-Weighted Pseudo-Regret

TACC

3,108.7444,249.1975,389.656,530.103May 8, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.05
3,277.7
2026.05
3,894
2026.05
4,023.4
2026.05
4,264.1
2026.05
4,317
2026.05
4,688.6
2026.05
5,083.2
2026.05
5,190.7
2026.05
5,201
2026.05
5,334.4
2026.05
5,359.1
2026.05
5,532.7
2026.05
6,668.6
2026.05
6,808.4
2026.05
6,827.5
2026.05
7,058.4
2026.05
7,331.9
2026.05
7,501.6