List your company

Instruction Following on IFEval (Normalized performance)

95.8Normalized performanceIso-CTS
35.16850.90966.6582.391Jun 15, 2026

Evaluation results

MethodLinks
Iso-CTSCategory=State-of-the-Art (+Many-shot), Backbone=Llama-3.2-3B2026.0695.8
ColD FusionCategory=Iterative / FL, Backbone=Llama-3.2-3B2026.0691.7
ConsensusTACategory=Conventional Method (+Many-shot), Backbone=Llama-3.2-3B2026.0691.7
METISCategory=Ours, Backbone=Llama-3.2-3B2026.0691.7
FedMergeCategory=Iterative / FL, Backbone=Llama-3.2-3B2026.0687.5
q-FedAvgCategory=Iterative / FL, Backbone=Llama-3.2-3B2026.0687.5
Iso-CCategory=State-of-the-Art (+Many-shot), Backbone=Llama-3.2-3B2026.0679.2
DARECategory=Conventional Method (+Many-shot), Backbone=Llama-3.2-3B2026.0679.2
TIESCategory=Conventional Method (+Many-shot), Backbone=Llama-3.2-3B2026.0679.2
TSV-MCategory=State-of-the-Art (+Many-shot), Backbone=Llama-3.2-3B2026.0675
TULU 3Category=Data-Mixing, Backbone=Llama-3.2-3B2026.0654.2
TA + Subspace BoostingCategory=State-of-the-Art (+Many-shot), Backbone=Llama-3.2-3B2026.0645.8
Task ArithmeticCategory=Conventional Method (+Many-shot), Backbone=Llama-3.2-3B2026.0637.5