Dan Su
NVIDIA research scientist working on LLM pretraining; co-lead author of the Nemotron-CC dataset paper
Dan Su is a research scientist at NVIDIA whose research focuses on large language model pretraining. Su is listed first among the co-first authors of the Nemotron-CC paper, which describes a 6.3-trillion-token English pretraining dataset built from Common Crawl, and was lead byline author of NVIDIA's announcement of the dataset.1234
Career
Su received a PhD in natural language processing from the Hong Kong University of Science and Technology and is a research scientist at NVIDIA, where Su's current research focuses on large language model pretraining.1
Nemotron-CC
Su, Kezhi Kong and Ying Lin are the co-first authors of "Nemotron-CC: Transforming Common Crawl into a Refined Long-Horizon Pretraining Dataset" (December 2024), published by NVIDIA's Applied Deep Learning Research group. The paper combines classifier ensembling, synthetic rephrasing of data and reduced reliance on heuristic filters to keep more high-quality tokens for long training runs.23
In January 2025 Su was the first byline author of NVIDIA's technical blog post announcing the dataset, which describes a curation pipeline built on NVIDIA NeMo Curator and the dataset's availability through Common Crawl.4
Work
Nemotron-CCOpen data
Co-first author of the Nemotron-CC paper and first byline author of NVIDIA's January 2025 blog post announcing the 6.3-trillion-token Common Crawl pretraining dataset.234
Career timeline
Related people
More sourced profiles at the same organization, in the same field, or from the same school.
- Aaron LefohnAlso at NVIDIA
- Adina WilliamsLanguage models
- Azita MartinAlso at NVIDIA
- Albert JiangLanguage models
- Bill DallyAlso at NVIDIA
- Alexander WettigLanguage models
- Boris GinsburgAlso at NVIDIA
- Ali GhodsiDatasets and data
- Bryan CatanzaroAlso at NVIDIA
- Andriy MulyarLanguage models
- David LuebkeAlso at NVIDIA
- Anton LozhkovLanguage models
Is this you?
Profiles cover public professional work only, and every line cites a source. If something is out of date, missing or wrong, send a correction with a link to an official page and we will update it. Suggest a correction
Add this badge to your website or GitHub profile README. It links back to this page. It is not an endorsement or a rating.
Markdown (for a README)
[](https://unitedstatesofamericasuperintelligence.com/people/dan-su/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/people/dan-su/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/profile.svg" alt="Dan Su — Profile on USASI" height="20"></a>
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.