Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Policy generalization on MESSENGER (S1)
Loading...
100,000
Win Rate
LED-WM
84,545.6
88,557.8
92,570
96,582.2
Nov 28, 2025
Win Rate
Updated 3mo ago
Evaluation Results
Method
Method
Links
Win Rate
LED-WM
Curriculum training=fa...
2025.11
100,000
CRL
Curriculum training=fa...
2025.11
88,250
EMMA (w/ curriculum)
Curriculum training=tr...
2025.11
88,230
EMMA (w/o curriculum)
Curriculum training=fa...
2025.11
85,140
Feedback
Search any
task
Search any
task