List your company

General Reasoning and Knowledge on Aggregate (BBH, MMLU, ARC-C, ThmQA)

26.27Mean ScoreHybKD
14.15417.299520.44523.5905May 25, 2026

Evaluation results

MethodLinks
HybKDDistillation Pair=Llama3.1-8B → Llama3.2-1B2026.0526.27
α-β divergenceDistillation Pair=Llama3.1-8B → Llama3.2-1B2026.0525.01
JS divergenceDistillation Pair=Llama3.1-8B → Llama3.2-1B2026.0524.84
Adaptive KLDistillation Pair=Llama3.1-8B → Llama3.2-1B2026.0524.41
Skew RKLDistillation Pair=Llama3.1-8B → Llama3.2-1B2026.0523.69
Skew FKLDistillation Pair=Llama3.1-8B → Llama3.2-1B2026.0523.37
Reverse KLDistillation Pair=Llama3.1-8B → Llama3.2-1B2026.0522.75
Total VariationDistillation Pair=Llama3.1-8B → Llama3.2-1B2026.0520.66
HybKDDistillation Pair=Gemma3-4B → Gemma3-1B2026.0516.26
Skew FKLDistillation Pair=Gemma3-4B → Gemma3-1B2026.0515.81
α-β divergenceDistillation Pair=Gemma3-4B → Gemma3-1B2026.0515.79
Reverse KLDistillation Pair=Gemma3-4B → Gemma3-1B2026.0515.78
JS divergenceDistillation Pair=Gemma3-4B → Gemma3-1B2026.0515.75
Skew RKLDistillation Pair=Gemma3-4B → Gemma3-1B2026.0515.08
Adaptive KLDistillation Pair=Gemma3-4B → Gemma3-1B2026.0514.63
Total VariationDistillation Pair=Gemma3-4B → Gemma3-1B2026.0514.62