List your company

Mathematical Reasoning on In-domain Reasoning Suite (AIME24, AIME25, AMC23, Math500, Minerva, Olympia) (test)

29.3AIME24 ScoreSFT+
2.269.2816.323.32Mar 2, 2026

Evaluation results

MethodLinks
SFT+Backbone=Qwen3-8b, Sampling=avg@16/avg@52026.0329.324.767.384.842.554.550.5
SPOTBackbone=Qwen3-8b, Sampling=avg@16/avg@52026.032827.371.587.439.758.552.1
RFTBackbone=Qwen3-8b, Sampling=avg@16/avg@52026.0327.320598340.453.847.3
Qwen3-8bSampling=avg@16/avg@52026.032219.366.582.837.952.346.8
SFTBackbone=Qwen3-8b, Sampling=avg@16/avg@52026.0315.313.35678.238.644.341
SPOTBackbone=Llama3.1-8b-Instruct, Sampling=avg@16/avg@52026.03422748.626.116.320.7
Llama3.1-8b-InstructSampling=avg@16/avg@52026.033.32.719.54722.416.918.6
SFTBackbone=Llama3.1-8b-Instruct, Sampling=avg@16/avg@52026.033.3221.546.819.515.118
RFTBackbone=Llama3.1-8b-Instruct, Sampling=avg@16/avg@52026.033.30244619.115.318
SFT+Backbone=Llama3.1-8b-Instruct, Sampling=avg@16/avg@52026.033.30.72648.625.715.319.9