List your company

Math Reasoning on AMC 2023 (avg@16, pass@16)

87.5Avg@16KPO-clipped
17.09235.37153.6571.929Feb 11, 2026Feb 28, 2026Mar 18, 2026Apr 4, 2026Apr 22, 2026May 9, 2026May 27, 2026

Evaluation results

MethodLinks
KPO-clippedClipping strategy=clipped, Number of candidate solutions=162026.0287.595
KPO-unclippedClipping strategy=unclipped, Number of candidate solutions=162026.028095
IRDSModel=QWEN3-4B, Data Budget=20%2026.0579.5—
GMPONumber of candidate solutions=162026.0276.5687.5
GSPONumber of candidate solutions=162026.0275.4695
GRPONumber of candidate solutions=162026.0273.4392.5
PPL-TOPModel=QWEN3-4B, Data Budget=20%2026.0572.4—
DEPOModel=QWEN3-4B, Data Budget=20%2026.0569—
IFDModel=QWEN3-4B, Data Budget=20%2026.0568.9—
LIMRModel=QWEN3-4B, Data Budget=20%2026.0568.6—
PPL-MIDDLEModel=QWEN3-4B, Data Budget=20%2026.0567.7—
RANDOMModel=QWEN3-4B, Data Budget=20%2026.0562.4—
IRDSModel=QWEN3-1.7B, Data Budget=20%2026.0555.8—
TOKEN-LENGTHModel=QWEN3-4B, Data Budget=20%2026.0555.2—
RANDOMModel=QWEN3-1.7B, Data Budget=20%2026.0548.9—
DEPOModel=QWEN3-1.7B, Data Budget=20%2026.0547.6—
LIMRModel=QWEN3-1.7B, Data Budget=20%2026.0545.1—
PPL-TOPModel=QWEN3-1.7B, Data Budget=20%2026.0545—
IFDModel=QWEN3-1.7B, Data Budget=20%2026.0543.2—
PPL-MIDDLEModel=QWEN3-1.7B, Data Budget=20%2026.0542.4—
TOKEN-LENGTHModel=QWEN3-1.7B, Data Budget=20%2026.0537.5—
TOKEN-LENGTHModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0523.1—
RANDOMModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0523—
DEPOModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0522.8—
LIMRModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0522.3—
IRDSModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0521.7—
PPL-TOPModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0520.7—
PPL-MIDDLEModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0520.4—
IFDModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0519.8—