Dani Yogatama
Co-founder and CEO of Reka and associate professor of computer science at USC
- Co-founder and CEO, Reka1
- Associate Professor, Department of Computer Science, University of Southern California1
Dani Yogatama is a co-founder and the CEO of Reka and an associate professor in the Department of Computer Science at the University of Southern California. Yogatama was previously a senior staff research scientist at DeepMind from 2016 to 2022, and is a co-author of the 2024 Reka technical report that introduced the Reka Core, Flash and Edge multimodal language models.12
Career
Yogatama received a PhD from Carnegie Mellon University (2010 to 2015), advised by Noah Smith as a member of Noah's ARK. Yogatama then worked at the Baidu Silicon Valley AI Lab (2015 to 2016) and at DeepMind (2016 to 2022), where Yogatama was a senior staff research scientist.1
Yogatama is now a co-founder and the CEO of Reka, and an associate professor of computer science at the University of Southern California.1
Research
Publications and news listed on Yogatama's site include work on semiparametric language models, random feature attention (ICLR 2021), episodic memory, and the cross-lingual transferability of monolingual representations (ACL 2020).13
At Reka, Yogatama co-authored the 2024 technical report on Reka Core, Flash and Edge, multimodal language models trained from scratch that take text, image, video and audio inputs.2
Work
Reka FlashPublished research
Co-author of "Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models" (2024), the Reka Team technical report describing the training and evaluation of the Reka Flash model and its siblings.2
Random Feature Attention (external site: arxiv.org)Published research
Co-author, with Hao Peng, Nikolaos Pappas, Roy Schwartz, Noah Smith and Lingpeng Kong, of the 2021 paper on random feature attention, which Yogatama's site notes was accepted to ICLR 2021 as a spotlight presentation.31
Career timeline
2010
Began a PhD at Carnegie Mellon University, advised by Noah Smith.1
2015
Completed the CMU PhD and joined the Baidu Silicon Valley AI Lab.1
2016
Joined DeepMind, later serving as a senior staff research scientist until 2022.1
Apr 2020
Had a paper on the cross-lingual transferability of monolingual representations accepted to ACL 2020.1
Mar 2021
Posted Random Feature Attention as a co-author; the paper was accepted to ICLR 2021 as a spotlight presentation.31
Apr 2024
Co-authored the Reka technical report introducing Reka Core, Flash and Edge.2
Other documented roles
Roles the sources document but that are not confirmed as current.
Related people
More sourced profiles at the same organization, in the same field, or from the same school.
- Abigail SeeAlso at Google DeepMind
- Alec RadfordLanguage models
- Abhimanyu DasAlso from USC
- Adam GleaveAlso at DeepMind
- Ali FarhadiAlso from Carnegie Mellon
- Albert GuAlso from Carnegie Mellon
- Alex GravesAlso at Google DeepMind
- Armen AghajanyanMultimodal models
- Alex SmolaAlso from Carnegie Mellon
- Anca DraganAlso at Google DeepMind
- Douwe KielaAlso at Google DeepMind
- Alexei A. EfrosAlso from Carnegie Mellon
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/dani-yogatama/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/people/dani-yogatama/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/profile.svg" alt="Dani Yogatama — 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.