List your company

Math Reasoning on AIME 2024 (mean@16)

42Mean@16IRDS
0.81611.50822.232.892May 27, 2026

Evaluation results

MethodLinks
IRDSModel=QWEN3-4B, Data Budget=20%2026.0542
LIMRModel=QWEN3-4B, Data Budget=20%2026.0530.6
PPL-TOPModel=QWEN3-4B, Data Budget=20%2026.0528.6
DEPOModel=QWEN3-4B, Data Budget=20%2026.0528.5
IFDModel=QWEN3-4B, Data Budget=20%2026.0528.4
PPL-MIDDLEModel=QWEN3-4B, Data Budget=20%2026.0527.9
IRDSModel=QWEN3-1.7B, Data Budget=20%2026.0523.3
RANDOMModel=QWEN3-4B, Data Budget=20%2026.0522.7
TOKEN-LENGTHModel=QWEN3-4B, Data Budget=20%2026.0518.4
LIMRModel=QWEN3-1.7B, Data Budget=20%2026.0517.2
DEPOModel=QWEN3-1.7B, Data Budget=20%2026.0514.5
RANDOMModel=QWEN3-1.7B, Data Budget=20%2026.0513.9
PPL-MIDDLEModel=QWEN3-1.7B, Data Budget=20%2026.0512.3
PPL-TOPModel=QWEN3-1.7B, Data Budget=20%2026.0512.2
IFDModel=QWEN3-1.7B, Data Budget=20%2026.0511.7
TOKEN-LENGTHModel=QWEN3-1.7B, Data Budget=20%2026.0510.7
DEPOModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.054.1
IRDSModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.053.4
IFDModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.053.2
TOKEN-LENGTHModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.053.1
PPL-MIDDLEModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.053.1
RANDOMModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.053
PPL-TOPModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.052.7
LIMRModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.052.4