Question Answering on SocialIQA (test)
78.1AccuracyExpert III (SocialIQA)
Evaluation Results
| Method | Links | |
|---|---|---|
| Expert III (SocialIQA)Seed Model=LLaMA-3.1-8B2026.05 | 78.1 | |
| MetaMoESeed Model=LLaMA-3.1-8B2026.05 | 77.64 | |
| BTMSeed Model=LLaMA-3.1-8B2026.05 | 77.28 | |
| ModelSoupSeed Model=LLaMA-3.1-8B2026.05 | 77.02 | |
| UnrestrictedMoESeed Model=LLaMA-3.1-8B2026.05 | 76.41 | |
| FlexOlmoSeed Model=LLaMA-3.1-8B2026.05 | 76.1 | |
| BTXSeed Model=LLaMA-3.1-8B2026.05 | 73.39 | |
| Expert I (CSQA)Seed Model=LLaMA-3.1-8B2026.05 | 69.19 | |
| Expert II (CosmosQA)Seed Model=LLaMA-3.1-8B2026.05 | 67.86 | |
| ZeroShotSeed Model=LLaMA-3.1-8B2026.05 | 57.57 |