Kezhi Kong
NVIDIA research scientist on the ADLR foundation model team; co-first author of Nemotron-CC
Kezhi Kong is a research scientist at NVIDIA on the foundation model team in Applied Deep Learning Research (ADLR), which trains the Nemotron models. Kong is a co-first author of the Nemotron-CC paper on a Common Crawl pretraining dataset, and earlier worked on graph representation learning and trustworthy machine learning during a PhD at the University of Maryland.123
Career
Kong earned a PhD in the Computer Science Department at the University of Maryland, College Park, advised by Tom Goldstein. Kong is now a research scientist at NVIDIA on the foundation model team in Applied Deep Learning Research, which trains the Nemotron family of models.1
Research
Kong's research covers large language models, graph representation learning and trustworthy machine learning. Earlier first-author papers include "Robust Optimization as Data Augmentation for Large-scale Graphs" (CVPR 2022), GOAT, a global transformer on large-scale graphs (ICML 2023), and OpenTab, on large language models as open-domain table reasoners (ICLR 2024); Kong also co-authored work on the reliability of watermarks for large language models (ICLR 2024).1
With Dan Su and Ying Lin, Kong is a co-first author of "Nemotron-CC: Transforming Common Crawl into a Refined Long-Horizon Pretraining Dataset" (December 2024), which uses classifier ensembling and synthetic rephrasing to build a large English pretraining dataset, and co-wrote NVIDIA's January 2025 announcement of the dataset.234
Work
Nemotron-CCOpen data
Co-first author of the Nemotron-CC paper and co-author of NVIDIA's blog post announcing the 6.3-trillion-token Common Crawl pretraining dataset.234
GOAT: A Global Transformer on Large-scale Graphs (external site: devnkong.github.io)Published research
First author of the ICML 2023 paper on a global transformer for large-scale graphs.1
Career timeline
2021
Co-first-authored VQ-GNN, a framework for scaling graph neural networks with vector quantization (NeurIPS 2021).1
2022
Published "Robust Optimization as Data Augmentation for Large-scale Graphs" as first author at CVPR 2022.1
2023
Published GOAT, a global transformer on large-scale graphs, as first author at ICML 2023.1
2024
Published OpenTab, on LLMs as open-domain table reasoners, as first author at ICLR 2024.1
Dec 3, 2024
Posted the Nemotron-CC paper as co-first author with Dan Su and Ying Lin.2
Jan 9, 2025
Co-authored NVIDIA's technical blog post announcing the Nemotron-CC dataset.4
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
- Dan SuAlso 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/kezhi-kong/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/people/kezhi-kong/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/profile.svg" alt="Kezhi Kong — 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.