Nemotron-CC-Math: A 133 Billion-Token-Scale High Quality Math Pretraining Dataset

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mahabadi, Rabeeh Karimi, Satheesh, Sanjeev, Prabhumoye, Shrimai, Patwary, Mostofa, Shoeybi, Mohammad, Catanzaro, Bryan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908496880992256
author Mahabadi, Rabeeh Karimi
Satheesh, Sanjeev
Prabhumoye, Shrimai
Patwary, Mostofa
Shoeybi, Mohammad
Catanzaro, Bryan
author_facet Mahabadi, Rabeeh Karimi
Satheesh, Sanjeev
Prabhumoye, Shrimai
Patwary, Mostofa
Shoeybi, Mohammad
Catanzaro, Bryan
contents Pretraining large language models (LLMs) on high-quality, structured data such as mathematics and code substantially enhances reasoning capabilities. However, existing math-focused datasets built from Common Crawl suffer from degraded quality due to brittle extraction heuristics, lossy HTML-to-text conversion, and the failure to reliably preserve mathematical structure. In this work, we introduce Nemotron-CC-Math, a large-scale, high-quality mathematical corpus constructed from Common Crawl using a novel, domain-agnostic pipeline specifically designed for robust scientific text extraction. Unlike previous efforts, our pipeline recovers math across various formats (e.g., MathJax, KaTeX, MathML) by leveraging layout-aware rendering with lynx and a targeted LLM-based cleaning stage. This approach preserves the structural integrity of equations and code blocks while removing boilerplate, standardizing notation into LaTeX representation, and correcting inconsistencies. We collected a large, high-quality math corpus, namely Nemotron-CC-Math-3+ (133B tokens) and Nemotron-CC-Math-4+ (52B tokens). Notably, Nemotron-CC-Math-4+ not only surpasses all prior open math datasets-including MegaMath, FineMath, and OpenWebMath-but also contains 5.5 times more tokens than FineMath-4+, which was previously the highest-quality math pretraining dataset. When used to pretrain a Nemotron-T 8B model, our corpus yields +4.8 to +12.6 gains on MATH and +4.6 to +14.3 gains on MBPP+ over strong baselines, while also improving general-domain performance on MMLU and MMLU-Stem. We present the first pipeline to reliably extract scientific content--including math--from noisy web-scale data, yielding measurable gains in math, code, and general reasoning, and setting a new state of the art among open math pretraining corpora. To support open-source efforts, we release our code and datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2508_15096
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Nemotron-CC-Math: A 133 Billion-Token-Scale High Quality Math Pretraining Dataset
Mahabadi, Rabeeh Karimi
Satheesh, Sanjeev
Prabhumoye, Shrimai
Patwary, Mostofa
Shoeybi, Mohammad
Catanzaro, Bryan
Computation and Language
Artificial Intelligence
Machine Learning
Pretraining large language models (LLMs) on high-quality, structured data such as mathematics and code substantially enhances reasoning capabilities. However, existing math-focused datasets built from Common Crawl suffer from degraded quality due to brittle extraction heuristics, lossy HTML-to-text conversion, and the failure to reliably preserve mathematical structure. In this work, we introduce Nemotron-CC-Math, a large-scale, high-quality mathematical corpus constructed from Common Crawl using a novel, domain-agnostic pipeline specifically designed for robust scientific text extraction. Unlike previous efforts, our pipeline recovers math across various formats (e.g., MathJax, KaTeX, MathML) by leveraging layout-aware rendering with lynx and a targeted LLM-based cleaning stage. This approach preserves the structural integrity of equations and code blocks while removing boilerplate, standardizing notation into LaTeX representation, and correcting inconsistencies. We collected a large, high-quality math corpus, namely Nemotron-CC-Math-3+ (133B tokens) and Nemotron-CC-Math-4+ (52B tokens). Notably, Nemotron-CC-Math-4+ not only surpasses all prior open math datasets-including MegaMath, FineMath, and OpenWebMath-but also contains 5.5 times more tokens than FineMath-4+, which was previously the highest-quality math pretraining dataset. When used to pretrain a Nemotron-T 8B model, our corpus yields +4.8 to +12.6 gains on MATH and +4.6 to +14.3 gains on MBPP+ over strong baselines, while also improving general-domain performance on MMLU and MMLU-Stem. We present the first pipeline to reliably extract scientific content--including math--from noisy web-scale data, yielding measurable gains in math, code, and general reasoning, and setting a new state of the art among open math pretraining corpora. To support open-source efforts, we release our code and datasets.
title Nemotron-CC-Math: A 133 Billion-Token-Scale High Quality Math Pretraining Dataset
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2508.15096