List your company

Open-ended MCQA on MMLU (16 subjects filtered)

97.11ASR@30REALISTA
-2.199623.582749.36575.1473May 12, 2026

Evaluation results

MethodLinks
REALISTATarget LLM=Llama-3-3B2026.0597.110.912.160.86
REALISTATarget LLM=Llama-3-8B2026.0593.61.331.153.48
ICDTarget LLM=Llama-3-3B2026.0590.771.5713.97100
ICDTarget LLM=Llama-3-8B2026.0587.321.7913.97100
LARGOTarget LLM=Llama-3-3B2026.0584.711.9341.0997.42
SECATarget LLM=Llama-3-8B2026.0582.971.720.292.59
SECATarget LLM=Llama-3-3B2026.0579.611.750.720.87
LARGOTarget LLM=Llama-3-8B2026.0557.922.6847.1196.45
Raw promptingTarget LLM=Llama-3-8B2026.0554.42.041.580
Raw promptingTarget LLM=Llama-3-3B2026.0545.481.921.580
REALISTATarget LLM=Qwen-2.5-7B2026.0541.612.753.322.88
SECATarget LLM=Qwen-2.5-7B2026.0532.472.321.153.76
SECATarget LLM=Qwen-2.5-14B2026.0527.512.260.583.48
REALISTATarget LLM=Qwen-2.5-14B2026.0527.242.490.873.17
LARGOTarget LLM=Qwen-2.5-7B2026.0523.892.2943.3796.53
ICDTarget LLM=Qwen-2.5-14B2026.0513.561.8413.97100
ICDTarget LLM=Qwen-2.5-7B2026.0511.51.7213.97100
LARGOTarget LLM=Qwen-2.5-14B2026.058.951.5138.3197.4
Raw promptingTarget LLM=Qwen-2.5-7B2026.056.40.991.580
Raw promptingTarget LLM=Qwen-2.5-14B2026.051.620.51.580