GRPO tuning on SLMs shows diminishing returns from hard math samples, with easier subsets matching full performance using 45% fewer steps and GSM8K training outperforming MATH training on numeric subsets.
Reading the remainders from bottom to top, we get629 10 =1165 8
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Limits of Difficulty Scaling: Hard Samples Yield Diminishing Returns in GRPO-Tuned SLMs
GRPO tuning on SLMs shows diminishing returns from hard math samples, with easier subsets matching full performance using 45% fewer steps and GSM8K training outperforming MATH training on numeric subsets.