List your company

Question Answering on SocialIQA (test)

78.1AccuracyExpert III (SocialIQA)
56.748862.291967.83573.3781May 14, 2026

Evaluation results

MethodLinks
Expert III (SocialIQA)Seed Model=LLaMA-3.1-8B2026.0578.1
MetaMoESeed Model=LLaMA-3.1-8B2026.0577.64
BTMSeed Model=LLaMA-3.1-8B2026.0577.28
ModelSoupSeed Model=LLaMA-3.1-8B2026.0577.02
UnrestrictedMoESeed Model=LLaMA-3.1-8B2026.0576.41
FlexOlmoSeed Model=LLaMA-3.1-8B2026.0576.1
BTXSeed Model=LLaMA-3.1-8B2026.0573.39
Expert I (CSQA)Seed Model=LLaMA-3.1-8B2026.0569.19
Expert II (CosmosQA)Seed Model=LLaMA-3.1-8B2026.0567.86
ZeroShotSeed Model=LLaMA-3.1-8B2026.0557.57