A controlled study of synthetic math data generation yields a new dataset blend (FLAMES) that improves fine-tuned 7B model scores on MATH and OlympiadBench, with caveats on checkpoint selection.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FLAMES: Improving LLM Math Reasoning via a Fine-Grained Analysis of the Data Synthesis Pipeline
A controlled study of synthetic math data generation yields a new dataset blend (FLAMES) that improves fine-tuned 7B model scores on MATH and OlympiadBench, with caveats on checkpoint selection.