List your company

Multilingual Understanding on M-HellaSwag

1.014PerformanceIso-CTS
0.674960.762980.8510.93902Jun 15, 2026

Evaluation results

MethodLinks
Iso-CTSCategory=State-of-the-Art (+Many-shot), Backbone=Llama-3.2-3B2026.061.014
Iso-CCategory=State-of-the-Art (+Many-shot), Backbone=Llama-3.2-3B2026.061.005
TSV-MCategory=State-of-the-Art (+Many-shot), Backbone=Llama-3.2-3B2026.061.005
DARECategory=Conventional Method (+Many-shot), Backbone=Llama-3.2-3B2026.061.005
FedMergeCategory=Iterative / FL, Backbone=Llama-3.2-3B2026.060.991
TULU 3Category=Data-Mixing, Backbone=Llama-3.2-3B2026.060.991
TIESCategory=Conventional Method (+Many-shot), Backbone=Llama-3.2-3B2026.060.991
METISCategory=Ours, Backbone=Llama-3.2-3B2026.060.991
ConsensusTACategory=Conventional Method (+Many-shot), Backbone=Llama-3.2-3B2026.060.981
ColD FusionCategory=Iterative / FL, Backbone=Llama-3.2-3B2026.060.977
TA + Subspace BoostingCategory=State-of-the-Art (+Many-shot), Backbone=Llama-3.2-3B2026.060.963
Task ArithmeticCategory=Conventional Method (+Many-shot), Backbone=Llama-3.2-3B2026.060.953
q-FedAvgCategory=Iterative / FL, Backbone=Llama-3.2-3B2026.060.688