Mathematical Reasoning on DeepMath 2025 (test)
55.8Pass@1FedAvg-PubSwap
Evaluation results
| Method | Links | |
|---|---|---|
| FedAvg-PubSwapModel=Qwen2-1.5B, Local steps (τ)=1202026.04 | 55.8 | |
| FedAvg-PubSwapModel=Qwen2-1.5B, Local steps (τ)=402026.04 | 54.6 | |
| FedAvg-PubSwapModel=Qwen2.5-Math-1.5B, Local steps (τ)=102026.04 | 54.5 | |
| FedAvg-PubSwapModel=Qwen2-1.5B, Local steps (τ)=102026.04 | 53.8 | |
| FedAvg-PubSwapModel=Qwen2-1.5B, Local steps (τ)=902026.04 | 53.3 | |
| FedAvg-GRPOModel=Qwen2.5-Math-1.5B, Local steps (τ)=102026.04 | 53.1 | |
| FedAvg-PubSwapModel=Qwen2.5-Math-1.5B, Local steps (τ)=402026.04 | 53.1 | |
| FedAvg-PubSwapModel=Qwen2.5-Math-1.5B, Local steps (τ)=1202026.04 | 53.1 | |
| FedAvg-GRPOModel=Qwen2-1.5B, Local steps (τ)=402026.04 | 52.7 | |
| FedProx-GRPOModel=Qwen2.5-Math-1.5B, Local steps (τ)=102026.04 | 52.6 | |
| FedProx-GRPOModel=Qwen2-1.5B, Local steps (τ)=402026.04 | 52.3 | |
| FedAvg-GRPOModel=Qwen2.5-Math-1.5B, Local steps (τ)=402026.04 | 52.3 | |
| FedProx-GRPOModel=Qwen2.5-Math-1.5B, Local steps (τ)=402026.04 | 52.1 | |
| FedProx-GRPOModel=Qwen2-1.5B, Local steps (τ)=102026.04 | 52 | |
| FedAvg-PubSwapModel=Qwen2.5-Math-1.5B, Local steps (τ)=902026.04 | 51 | |
| FedAvg-GRPOModel=Qwen2-1.5B, Local steps (τ)=1202026.04 | 50.7 | |
| FedAvg-GRPOModel=Qwen2.5-Math-1.5B, Local steps (τ)=902026.04 | 50.5 | |
| FedAvg-GRPOModel=Qwen2-1.5B, Local steps (τ)=902026.04 | 50.4 | |
| FedAvg-GRPOModel=Qwen2-1.5B, Local steps (τ)=102026.04 | 49.5 | |
| FedProx-GRPOModel=Qwen2.5-Math-1.5B, Local steps (τ)=1202026.04 | 49.4 | |
| FedAvg-GRPOModel=Qwen2.5-Math-1.5B, Local steps (τ)=1202026.04 | 49.3 | |
| FedProx-GRPOModel=Qwen2-1.5B, Local steps (τ)=902026.04 | 48 | |
| FedProx-GRPOModel=Qwen2.5-Math-1.5B, Local steps (τ)=902026.04 | 48 | |
| FedProx-GRPOModel=Qwen2-1.5B, Local steps (τ)=1202026.04 | 47.7 | |
| Base modelModel=Qwen2.5-Math-1.5B, Local steps (τ)=102026.04 | 34.9 | |
| Base modelModel=Qwen2.5-Math-1.5B, Local steps (τ)=402026.04 | 34.9 | |
| Base modelModel=Qwen2.5-Math-1.5B, Local steps (τ)=902026.04 | 34.9 | |
| Base modelModel=Qwen2.5-Math-1.5B, Local steps (τ)=1202026.04 | 34.9 | |
| Base modelModel=Qwen2-1.5B, Local steps (τ)=102026.04 | 14.9 | |
| Base modelModel=Qwen2-1.5B, Local steps (τ)=402026.04 | 14.9 | |
| Base modelModel=Qwen2-1.5B, Local steps (τ)=902026.04 | 14.9 | |
| Base modelModel=Qwen2-1.5B, Local steps (τ)=1202026.04 | 14.9 |