Accuracy on Brumo 2025 (Mathematical Reasoning)
54.79AccuracyDelTA
Evaluation Results
| Method | Links | |
|---|---|---|
| DelTABackbone=Qwen3-14B-Base, Fine-tuning (FT)=false2026.05 | 54.79 | |
| DAPOBackbone=Qwen3-14B-Base, Fine-tuning (FT)=true2026.05 | 52.08 | |
| FIPOBackbone=Qwen3-14B-Base, Fine-tuning (FT)=false2026.05 | 52.08 | |
| SAPOBackbone=Qwen3-14B-Base, Fine-tuning (FT)=false2026.05 | 50.21 | |
| DAPOBackbone=Qwen3-14B-Base, Fine-tuning (FT)=false2026.05 | 48.13 | |
| DelTABackbone=Qwen3-8B-Base, Fine-tuning (FT)=false2026.05 | 44.79 | |
| SAPOBackbone=Qwen3-8B-Base, Fine-tuning (FT)=false2026.05 | 39.37 | |
| DAPOBackbone=Qwen3-8B-Base, Fine-tuning (FT)=true2026.05 | 39.17 | |
| FIPOBackbone=Qwen3-8B-Base, Fine-tuning (FT)=false2026.05 | 37.71 | |
| DAPOBackbone=Qwen3-8B-Base, Fine-tuning (FT)=false2026.05 | 36.46 |