List your company

Algorithmic Reasoning on BBEH Mini

17.8AccuracyStatic
13.53614.64315.7516.857Apr 6, 2026

Evaluation results

MethodLinks
StaticBudget=40962026.0417.812,428
TABBudget (B)=10k2026.04175,236
StaticBudget=20482026.0416.56,984
TABBudget (B)=8k2026.0415.74,644
StaticBudget=10242026.0415.44,394
TABBudget (B)=5k2026.0415.43,867
StaticBudget=5122026.0414.62,909
TABBudget (B)=3k2026.0414.62,238
LLM-Judge IndividualSelection Strategy=Individual2026.0413.92,152
LLM-Judge Multi-TurnSelection Strategy=Multi-Turn2026.0413.91,949
StaticBudget=2562026.0413.72,054