List your company

Multiple-choice Question Answering on MMLU Stem (val)

58.13AccuracySP-Struct-RWD1-full
50.64252.58654.5356.474Dec 8, 2025

Evaluation results

MethodLinks
SP-Struct-RWD1-fullInference Protocol=Agentic-Independent, Prompting=zero-shot, Number of Parameters=3B2025.1258.13
SP-Struct-RWD1-fullInference Protocol=Agentic-Internal, Prompting=zero-shot, Number of Parameters=3B2025.1256.8
DeepSeekMath Base-7BPrompting=few-shot, Number of Parameters=7B2025.1256.5
SP-Struct-RWD1-fullInference Protocol=Multiple-Try, Prompting=zero-shot, Number of Parameters=3B2025.1253.6
Mistral 7BPrompting=5-shot, Number of Parameters=7B2025.1251.1
SP-Struct-RWD1-fullInference Protocol=Single-Try, Prompting=zero-shot, Number of Parameters=3B2025.1250.93