Skip to main content
Marco Baroni

Marco Baroni

Universitat Pompeu Fabra

Social & Behavioural Sciences

Marco Baroni received a PhD in Linguistics from the University of California, Los Angeles, in the year 2000. After several experiences in research and industry, he joined the Center for Mind/Brain Sciences of the University of Trento, where he became associate professor in 2013. In 2016, Marco joined the Facebook Artificial Intelligence Research team. In 2019, he became ICREA research professor, affiliated with the Linguistics Department of Pompeu Fabra University in Barcelona. Marco's work in the areas of multimodal and compositional distributed semantics has received widespread recognition, including a Google Research Award, an ERC Starting Grant, and the ICAI-JAIR best paper prize and the ACL test of time award. Marco's current research focuses on emergent communication in communities of deep-learning-trained artificial neural networks. In 2021, he was awarded an ERC Advanced Grant to work on this topic.

Research interests

Current deep-learning-trained artificial neural networks (“deep nets”) have revolutionized AI, but they are highly specialized devices, each optimized to solve a specific task. Taking inspiration from communities of human experts, that can solve problems together by collaborating through language, I want to teach specialized deep nets to communicate with each other in order to solve complex tasks. The key insight is that it would be hopeless to specify an exhaustive communication protocol for deep nets by hand. By relying on their strong learning capabilities, I let them instead evolve their own protocol by exposing them to joint tasks which they can only solve by cooperating through communication, and I study the characteristics of this emergent protocol.

Selected publications

– Lakretz Y, Hupkes D, Vergallito A, Marelli M, Baroni M & Dehaene S 2021, ‘Mechanisms for handling nested dependencies in neural-network language models and humans’, Cognition, 213, 104699.

– Chaabouni R, Kharitonov E, Dupoux E, & Baroni M 2021, ‘Communicating artificial neural networks develop efficient color-naming systems.’ Proceedings of the National Academy of Science (PNAS), 18 (12) e2016569118.

– Linzen T & Baroni M 2021. ‘Syntactic Structure from Deep Learning’. Annual Review of Linguistics 7: 195-212.

Selected research activities

– Keynote at EACL 2021

– Invited talk at the Collège de France

– Area chair at  ICLR 2021


ICREA Memoir 2021