List your company

Mathematical Reasoning on AIME 25 (pass@1, pass@256)

13Pass@1F-CISPO
0.2083.5296.8510.171Feb 6, 2026

Evaluation results

MethodLinks
F-CISPOBackbone=Qwen2.5-7B, Group Size (N)=82026.021364.6
DAPOBackbone=Qwen2.5-7B, Group Size (N)=82026.021245.6
F-DAPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0211.552.9
F-CISPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0211.259.7
F-GRPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0210.557.9
CISPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0210.553.8
F-DAPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.0210.361
F-GRPOBackbone=Qwen2.5-7B, Group Size (N)=82026.0210.152.6
GRPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.029.958
DAPOBackbone=Qwen2.5-1.5B-Math, Group Size (N)=82026.029.859.2
CISPOBackbone=Qwen2.5-7B, Group Size (N)=82026.029.759.8
GRPOBackbone=Qwen2.5-7B, Group Size (N)=82026.026.740.8
F-CISPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.02224.5
F-DAPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.021.728.7
F-GRPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.02129.5
DAPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.02118.5
CISPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.02125.4
GRPOBackbone=Llama-3.2-3B-Instruct, Group Size (N)=82026.020.721.5