Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Reward Hacking Analysis on Reward Hacking Evaluation Environment
Loading...
148.84
V^pi
EO
43.4672
70.8236
98.18
125.5364
Apr 5, 2026
V^pi
V^pi / V^*
||v||_w
v_3 (C3)
RHSI
Updated 1mo ago
Evaluation Results
Method
Method
Links
V^pi
V^pi / V^*
||v||_w
v_3 (C3)
RHSI
EO
Condition=EO
2026.04
148.84
100
64.5
70.8
64.5
MO
Condition=MO
2026.04
77.95
52.4
60.6
58
31.7
SmartTutor
Condition=ST
2026.04
64.42
43.3
23.7
5.8
10.2
MAS
Condition=MAS
2026.04
47.52
31.9
32.8
39.1
10.5
Feedback
Search any
task
Search any
task