List your company

Multiple Choice Question Answering on ARC Challenge o.o.d.

90.65AccuracyWALLA I
65.648472.139278.6385.1208Jul 5, 2026

Evaluation results

MethodLinks
WALLA Ipooling=log-pool2026.0790.652.9679.9430.419.16
RouteLLM2026.0790.612.3789.1846.389.35
Gemma22026.0790.437.5589.1838.4310.67
StackedGen2026.0790.352.1989.0446.058.85
WALLA II2026.0790.286.1788.0742.2310.59
NIRTRouter2026.0790.175.1289.1846.3810.65
WALLA I2026.0789.896.2486.7644.911.06
WALLA Icalibration=calibrated2026.0789.598.6478.7750.249.84
SelfCertainty2026.0789.222.791.5966.9610.71
RouterDC2026.0787.487.2263.2127.513.98
UniformAvg2026.0787.397.4—014.23
PackLLM2026.0786.266.8736.38-21.814.93
Llama3.12026.0781.33.0637.43-12.221.01
Aloe2026.0780.358.5851.17-0.4323.43
BioMistral2026.0766.6110.9530.68-25.840.65