List your company

Mathematical Reasoning on Mathematical Reasoning Aggregate Average of 6 Benchmarks (Mean@8)

65.5Mean@8 ScoreDISA
21.71633.08344.4555.817May 17, 2026

Evaluation results

MethodLinks
DISABackbone=Qwen3-4B-Base2026.0565.5
FlowRLBackbone=Qwen3-4B-Base2026.0564.8
GSPOBackbone=Qwen3-4B-Base2026.0562.7
GRPOBackbone=Qwen3-4B-Base2026.0549.4
DISABackbone=Qwen2.5-7B2026.0533.5
FlowRLBackbone=Qwen2.5-7B2026.0533.2
GRPOBackbone=Qwen2.5-7B2026.0532.8
VanillaBackbone=Qwen3-4B-Base2026.0529.7
GSPOBackbone=Qwen2.5-7B2026.0529.3
VanillaBackbone=Qwen2.5-7B2026.0523.4