In this talk I present two novel approaches, Generationary and Exemplification Modeling, that go beyond the mainstream assumption that word senses can be represented as discrete items of a predefined inventory, and put forward generative models which produce contextualized definitions and usage examples for arbitrary lexical items, from words to phrases. Generationary employs a novel span-based encoding scheme to fine-tune an English pre-trained Encoder-Decoder system and generate new definitions. Generationary outperforms previous approaches in the generative task of Definition Modeling in many settings, but it also matches or surpasses the state of the art in discriminative tasks such as Word Sense Disambiguation and Word-in-Context. Exemplification Modeling takes the opposite perspective: it puts forward a novel task and a seq2seq architecture which generates usage examples given one or more words with their sense definitions. In addition to their considerable degree of freedom in "understanding" lexical meanings, we show that both approaches benefit from training on definitions from multiple inventories, with strong gains across benchmarks.
Roberto Navigli is Professor at the Sapienza University of Rome, where he leads the Sapienza NLP Group. He has received two ERC grants on multilingual word sense disambiguation (2011-2016) and multilingual language- and syntax-independent open-text unified representations (2017-2022). In 2015 he received the META prize for groundbreaking work in overcoming language barriers with BabelNet, a project also highlighted in The Guardian and Time magazine, and winner of the Artificial Intelligence Journal prominent paper award 2017. He is the co-founder of Babelscape, a successful deep tech company which enables Natural Language Understanding in dozens of languages. He is a Program Co-Chair of ACL-IJCNLP 2021.