Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Temporal Regret Control on CAP-GSM8K E=500 Llama-3.3-70b
Loading...
2.4
RCE_temp(E)
Trivium
-7.624
60.038
127.7
195.362
Jun 3, 2026
RCE_temp(E)
Alpha Hat (α^)
Commit Rate
Verdict Margin
Updated 1mo ago
Evaluation Results
Method
Method
Links
RCE_temp(E)
Alpha Hat (α^)
Commit Rate
Verdict Margin
Trivium
Model=Llama-3.3-70b, H...
2026.06
2.4
0.39
100
10
RLVR
Model=Llama-3.3-70b, H...
2026.06
248.8
40
0
-
Reactive
Model=Llama-3.3-70b, H...
2026.06
253
40.7
0
-
Feedback
Search any
task
Search any
task
