Google DeepMind Academic Fellow to advance multilingual AI research at UCL 

Dr Tonja has been appointed the second Google DeepMind Academic Fellow at UCL’s Centre for Artificial Intelligence, succeeding David Ifeoluwa Adelani.  

Having received a PhD in Computer Science from the Instituto Politécnico Nacional in Mexico, Dr Tonja went on to work as a Postdoctoral Researcher at the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) in Abu Dhabi, UAE, where he specialised in natural language processing and multilingual AI.  

First established at UCL in 2022, the Google DeepMind Academic Fellowship Programme aims to support this kind of leading-edge postdoctoral research by providing opportunities to early-career researchers which develop their academic and leadership skills.  

Dr Tonja’s focus is now on multilingual and under-resourced language processing, including the development and evaluation of large language models tailored for low-resource languages. His work also explores culturally relevant evaluation benchmarks, cross-lingual transfer learning, and multimodal approaches that combine speech, text and vision-language models. 

Published in leading venues such as ACL, NAACL, EMNLP, NeurIPS, and TACL, Dr Tonja has earned wide recognition in his field. In 2024, he received the EMNLP Outstanding Paper Award and led the development of Afri-MCQA, a multilingual African cultural question-answering benchmark, and InkubaLM, an African-focused small language model. 

“With Atnafu, we are delighted to welcome another Google DeepMind Fellow to UCL. He already has a strong track record, and we look forward to working with him to further his research and career as part of the UCL NLP Group and the wider AI Centre,’’ said Pontus Stenetorp, head of the group and Professor of Natural Language Processing.

Google DeepMind Academic Fellowship 

Dr Tonja’s appointment marks the latest chapter in a unique partnership between UCL and Google DeepMind. In the last decade, Google DeepMind has supported more than forty MSc scholars, six PhD candidates in the UKRI Foundational AI CDT, a Google DeepMind Chair of Machine Learning and Artificial Intelligence, and the creation of the initial Google DeepMind Academic Fellowship in 2022. 

Dr Atnafu’s arrival coincides with the launch of the Here, it will happen campaign at UCL, which will convene far-sighted partners and philanthropists to develop and harness technologies with the potential to transform how we live, work and think about the world around us. 

Strengthening AI research at UCL 

The AI Centre at UCL is a major research unit focused on foundational AI, developing next-generation techniques across natural language processing, the natural sciences, robotics and related areas. 

Sustainable development of AI is central to its mission, including supporting the expansion of AI technologies to underrepresented sectors. The continuing partnership with Google DeepMind strengthens this work by attracting leading research talent and supporting UCL’s broader ambition to deliver global impact. 
  
“Supporting the next generation of high-potential talent to accelerate key research challenges using AI is central to the Google DeepMind Academic Fellowship Programme,” said Melanie Eusebe, Director of AI and Society at the Google DeepMind Impact Accelerator. “We are proud to fund UCL’s efforts to bring together world-leading expertise with an impact focus, enabling research with the potential to widen access to the benefits of AI for many more people globally.”     

“Dr Atnafu’s appointment is a powerful reminder of what is possible when world-leading research ambition is matched by a deep and visionary partnership,” said Angharad Milenkovic, Vice-President (Advancement) at UCL.  “His work to build AI that works for linguistically diverse and underrepresented communities is exactly the kind of research that underpins the Here, it will happen campaign. 

“We are grateful to Google DeepMind for their sustained commitment to this Fellowship, and to the broader mission of unlocking the transformational potential of AI for the benefit of us all.”       

About Google DeepMind       

Google DeepMind is a world-leading AI research lab with British heritage and an international team, committed to building AI responsibly, delivering scientific breakthroughs, and creating products that improve billions of lives.  

Its breakthroughs over the last decade include AlphaGo — the first computer program to defeat a Go world champion — Transformers, the neural networks that underpin all modern language models, and AlphaFold, an AI model that can accurately predict the structure of proteins, DNA, RNA, ligands and more, and Gemini, a family of versatile AI models built from the ground up for multimodality, seamlessly combining and understanding text, code, images, audio and video.

Image 

Dr Atnafu Lambebo Tonja.

Further links

The Google DeepMind Academic Fellowship Programme 
UCL Centre for Artificial Intelligence
Dr Atnafu Lambebo Tonja on Github