Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Physical Commonsense Reasoning on PIQA (Mean per-step regret)
Loading...
0.152
Mean Per-Step Regret
LinFTPL
0.14448
0.19524
0.246
0.29676
Feb 23, 2026
Mean Per-Step Regret
Updated 4d ago
Evaluation Results
Method
Method
Links
Mean Per-Step Regret
LinFTPL
Strategy Category=Cont...
2026.02
0.152
Simpl
Strategy Category=Stat...
2026.02
0.161
No-Rewrite (NoRw)
Strategy Category=Base
2026.02
0.172
LinEXP3
Strategy Category=Cont...
2026.02
0.173
Para
Strategy Category=Stat...
2026.02
0.174
TS
Strategy Category=Non-...
2026.02
0.174
TS
Strategy Category=Cont...
2026.02
0.186
ϵ-FTRL
Strategy Category=Non-...
2026.02
0.192
LinUCB+KL
Strategy Category=Cont...
2026.02
0.193
LinUCB
Strategy Category=Cont...
2026.02
0.197
EXP3
Strategy Category=Non-...
2026.02
0.213
Clarify
Strategy Category=Stat...
2026.02
0.236
Disamb
Strategy Category=Stat...
2026.02
0.252
FTPL
Strategy Category=Non-...
2026.02
0.259
Expand
Strategy Category=Stat...
2026.02
0.34
Feedback
Search any
task
Search any
task