List your company

Mathematical Reasoning on Olympiad (pass@1, pass@256)

42.4Pass@1F-CISPO
18.4824.6930.937.11Feb 6, 2026

Evaluation results

MethodLinks
F-CISPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0242.477.5
CISPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0241.576.9
F-DAPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0240.975.6
DAPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0240.571.8
GRPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0237.865.8
CISPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0237.671.9
F-GRPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0237.571.6
F-DAPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0237.276.2
F-GRPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.023775.5
DAPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0236.473.2
F-CISPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0236.474
GRPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0236.374.3
F-DAPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0223.261.3
CISPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0222.555.4
F-CISPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0221.558.7
DAPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.022148.4
F-GRPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0220.161.3
GRPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.0219.459.3