Generative AI is exploding: numerous large language models (LLMs), and even large multimodal models, are now available, both for research and application purposes. Several established machine learning practices are now being rapidly replaced by other practices (e.g., few-shot prompting, instruction-tuning), for which we do not yet have clear guidelines. In the talk I will provide a personal review of some features of LLMs that I have found interesting and useful (given my background and research interests) and some drawbacks that, from my point of view, are still affecting LLMs.
Speaker: Bernardo Magnini
Bernardo Magnini is senior researcher at Fondazione Bruno Kessler – FBK (Trento, Italy), and Director of the FBK International PhD Program. His interests are in the field of Computational Linguistics, particularly lexical semantics and lexical resources, question answering, textual entailment, and conversational agents.
He has launched EVALITA, the evaluation campaign for both NLP and speech tools for the Italian language, has co-chaired CLIC-it 2014 (the first Italian conference on Computational Linguistics), AI*IA 2018 and CLiC-it 2023. He has been local organizer of EACL 2006 in Trento and of ACL 2019 in Florence, and General Chair of ACL 2022, the 60th Annual Meeting of the Association for Computational Linguistics. He has been President of the Italian Italian Association for Computational Linguistics (AILC) from 2015 to 2022.