List your company

General performance assessment on Overall Combined Benchmarks

49.64Performance (Seen Data)GRPO
22.454429.512236.5743.6278May 25, 2026

Evaluation results

MethodLinks
GRPOBackbone=Gemma3-4b-it2026.0549.6438.8346.67
LRPOBackbone=Gemma3-4b-it2026.0549.5139.8646.89
DPOBackbone=Gemma3-4b-it2026.0548.938.0546
MAPOBackbone=Gemma3-4b-it2026.0548.7638.0845.86
VanillaBackbone=Gemma3-4b-it2026.0548.6738.2345.8
LIDRBackbone=Gemma3-4b-it2026.0548.2637.4645.29
MPOBackbone=Gemma3-4b-it2026.0548.0437.8245.19
LRPOBackbone=Qwen2.5-1.5b-it2026.0537.9419.0832.15
GRPOBackbone=Qwen2.5-1.5b-it2026.0535.0919.4430.42
DPOBackbone=Qwen2.5-1.5b-it2026.0533.5519.2229.33
VanillaBackbone=Qwen2.5-1.5b-it2026.0532.8618.5428.64
MAPOBackbone=Qwen2.5-1.5b-it2026.0532.4718.5228.4
MPOBackbone=Qwen2.5-1.5b-it2026.0532.3318.7928.38
LIDRBackbone=Qwen2.5-1.5b-it2026.0531.6517.9727.58
LRPOBackbone=LLaMA3.2-1b-it2026.0524.2718.4822.29
MPOBackbone=LLaMA3.2-1b-it2026.0524.0917.6121.86
GRPOBackbone=LLaMA3.2-1b-it2026.052418.422.18
LIDRBackbone=LLaMA3.2-1b-it2026.0523.9717.5221.69
DPOBackbone=LLaMA3.2-1b-it2026.0523.8118.1921.76
MAPOBackbone=LLaMA3.2-1b-it2026.0523.7417.8621.64
VanillaBackbone=LLaMA3.2-1b-it2026.0523.517.2221.36