Recent work from DeepMind on grounded language learning in virtual environments:

Emergent Systematic Generalization in a Situated AgentFelix Hill, Andrew Lampinen, Rosalia Schneider, Stephen Clark, Matthew Botvinick, James L. McClelland, Adam Santoro:

Understanding Early Word Learning in Situated Artificial AgentsFelix Hill, Stephen Clark, Karl Moritz Hermann, Phil Blunsom:

These are the academic web pages of Stephen Clark, who is now a full-time Research Scientist at DeepMind, and an Honorary Professor at Queen Mary University of London. Previously he spent 18 years working at UK universities, first as a postdoctoral researcher at the University of Edinburgh; then as a member of Faculty at the Oxford University Department of Computer Science, and a Fellow of Keble College, Oxford; and finally as a member of Faculty at the University of Cambridge Department of Computer Science and Technology. He holds a PhD in Computer Science and Artificial Intelligence from the University of Sussex, and an MA in Philosophy from Gonville and Caius College, Cambridge. He has been the recipient of ERC Consolidator and  Proof-of-Concept Grants (2012-2018), Chair of the European Chapter of the Association for Computational Linguistics (2011-2015), and program co-chair for the flagship ACL conference (in 2010). He is an Associate Editor for the Journal of Artificial Intelligence Research, and has been an Action Editor for the Transactions of the Association for Computational Linguistics, as well as on the Editorial Boards of Computational Linguistics, Journal of Artificial Intelligence Research, Computer Speech and Language, and the Journal of Natural Language Engineering. He carries out research at the intersection of Natural Language Processing, Computational Linguistics and Machine Learning, with much of his previous work focusing on the syntactic and semantic analysis of natural language text. His current research focus is the acquisition of language by artificial agents in the context of realistic virtual environments.