List your company

Mathematical Reasoning on AMC23 (Avg@16 and Pass@16)

53.75Average Score @16PriFT-mass
20.927629.448837.9746.4912Jun 8, 2026

Evaluation results

MethodLinks
PriFT-massBackbone=Qwen2.5-Math (7B)2026.0653.7590
PriFT-probBackbone=Qwen2.5-Math (7B)2026.0649.8490
PriFT-massBackbone=Qwen3-Base (8B)2026.0647.0385
IDFTBackbone=Qwen2.5-Math (7B)2026.0645.3177.5
PriFT-probBackbone=Qwen3-Base (8B)2026.0644.0682.5
DFTBackbone=Qwen3-Base (8B)2026.0643.5977.5
IDFTBackbone=Qwen3-Base (8B)2026.064367.5
ASFTBackbone=Qwen2.5-Math (7B)2026.0642.577.5
TALRBackbone=Qwen2.5-Math (7B)2026.0642.0380
DFTBackbone=Qwen2.5-Math (7B)2026.0640.9472.5
TALRBackbone=Qwen3-Base (8B)2026.0640.982.5
ASFTBackbone=Qwen3-Base (8B)2026.0639.0685
OriginalBackbone=Qwen3-Base (8B)2026.0635.4785
SFTBackbone=Qwen2.5-Math (7B)2026.062570
OriginalBackbone=Qwen2.5-Math (7B)2026.0624.8482.5
EAFTBackbone=Qwen3-Base (8B)2026.0624.270
SFTBackbone=Qwen3-Base (8B)2026.0622.8162.5
EAFTBackbone=Qwen2.5-Math (7B)2026.0622.1975