ResearchBenchmarksMathematical Reasoning on MATH 500 (mean@16)Follow89Mean@16 AccuracyIRDS53.1262.43571.7581.065May 27, 2026Evaluation ResultsMethodMethodLinksMean@16 AccuracyIRDSModel=QWEN3-4B, Data B...Model=QWEN3-4B, Data Budget=10%2026.0589IRDSModel=QWEN3-1.7B, Data...Model=QWEN3-1.7B, Data Budget=10%2026.0580.5IRDSModel=LLAMA-3.1-8B-INS...Model=LLAMA-3.1-8B-INSTRUCT, Data Budget=10%2026.0554.5