List your company

Math Reasoning on Minerva (mean@16)

43.8Mean@16IRDS
23.5228.78534.0539.315May 27, 2026

Evaluation results

MethodLinks
IRDSModel=QWEN3-4B, Data Budget=20%2026.0543.8
PPL-MIDDLEModel=QWEN3-4B, Data Budget=20%2026.0542.9
LIMRModel=QWEN3-4B, Data Budget=20%2026.0542.7
DEPOModel=QWEN3-4B, Data Budget=20%2026.0541.8
PPL-TOPModel=QWEN3-4B, Data Budget=20%2026.0540.8
IFDModel=QWEN3-4B, Data Budget=20%2026.0540.3
RANDOMModel=QWEN3-4B, Data Budget=20%2026.0539.4
TOKEN-LENGTHModel=QWEN3-4B, Data Budget=20%2026.0538.9
IRDSModel=QWEN3-1.7B, Data Budget=20%2026.0533.6
LIMRModel=QWEN3-1.7B, Data Budget=20%2026.0531.7
DEPOModel=QWEN3-1.7B, Data Budget=20%2026.0531.1
RANDOMModel=QWEN3-1.7B, Data Budget=20%2026.0530
PPL-TOPModel=QWEN3-1.7B, Data Budget=20%2026.0529.8
PPL-MIDDLEModel=QWEN3-1.7B, Data Budget=20%2026.0529.5
IFDModel=QWEN3-1.7B, Data Budget=20%2026.0528.3
IRDSModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0528.3
DEPOModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0527.4
TOKEN-LENGTHModel=QWEN3-1.7B, Data Budget=20%2026.0527.2
PPL-MIDDLEModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0525.9
LIMRModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0525.7
RANDOMModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0525.3
TOKEN-LENGTHModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0525.2
IFDModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0525
PPL-TOPModel=LLAMA-3.1-8B-INSTRUCT, Data Budget=20%2026.0524.3