OpenThoughts3-7B, trained on a 1.2M-example public dataset via systematic pipeline optimizations, achieves 53% on AIME 2025, 51% on LiveCodeBench, and 54% on GPQA Diamond.
Find the sum of the squares of its coefficients
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
-
OpenThoughts: Data Recipes for Reasoning Models
OpenThoughts3-7B, trained on a 1.2M-example public dataset via systematic pipeline optimizations, achieves 53% on AIME 2025, 51% on LiveCodeBench, and 54% on GPQA Diamond.