List your company

Math Reasoning on MATH 500 (mean@16)

91.3Mean@16 AccuracyIRDS
50.42861.03971.6582.261May 27, 2026

Evaluation results

MethodLinks
IRDSModel=QWEN3-4B, Data Budget=20%2026.0591.3
LIMRModel=QWEN3-4B, Data Budget=20%2026.0589.1
PPL-MIDDLEModel=QWEN3-4B, Data Budget=20%2026.0588.9
DEPOModel=QWEN3-4B, Data Budget=20%2026.0588.8
PPL-TOPModel=QWEN3-4B, Data Budget=20%2026.0587.3
IFDModel=QWEN3-4B, Data Budget=20%2026.0586.2
TOKEN-LENGTHModel=QWEN3-4B, Data Budget=20%2026.0585.3
RANDOMModel=QWEN3-4B, Data Budget=20%2026.0584.5
IRDSModel=QWEN3-1.7B, Data Budget=20%2026.0582.5
LIMRModel=QWEN3-1.7B, Data Budget=20%2026.0579
DEPOModel=QWEN3-1.7B, Data Budget=20%2026.0577.1
PPL-MIDDLEModel=QWEN3-1.7B, Data Budget=20%2026.0575.9
TOKEN-LENGTHModel=QWEN3-1.7B, Data Budget=20%2026.0574.8
IFDModel=QWEN3-1.7B, Data Budget=20%2026.0574.1
PPL-TOPModel=QWEN3-1.7B, Data Budget=20%2026.0573.9
RANDOMModel=QWEN3-1.7B, Data Budget=20%2026.0572.6
IRDSModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0555
RANDOMModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0554.6
DEPOModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0554
TOKEN-LENGTHModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0553.3
PPL-TOPModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0553.1
IFDModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0553
LIMRModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0552.6
PPL-MIDDLEModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0552