Keynotes Speakers

Bio: Slav Petrov is Vice President, Research at Google DeepMind. He is a co-lead on Gemini, Google’s large models effort. His work has been recognized with multiple Best Paper Awards (ACL’11, NAACL’12, ACL’16, 10-year Test-of-Time Award at ACL’23) and provides better language capabilities to billions of users in a variety of Google products spanning Web Search, Assistant, Ads, Translate & Cloud. Slav is the recipient of the 2014 John Atanasoff Award by the President of Bulgaria and a World Champion at RoboCup 2004. For many years, Slav taught Statistical Natural Language Processing at New York University. He holds a PhD from the University of California at Berkeley.

Slav has spent roughly equal parts of his life in Bulgaria, Germany and the US. Whenever Bulgaria plays Germany in soccer, he supports Bulgaria.

Abstract: Drawing upon my 25-year research journey, this talk offers a perspective on the current state of Artificial Intelligence (AI), highlighting both its remarkable progress and the significant challenges that remain. In recent years, AI has taken giant leaps, fueled by key developments in deep learning, transformer models, and scaling. I will briefly review this progress and then turn our attention to the open challenges we are still facing. Despite the recent advancements, AI is still in its infancy. All AI systems still struggle with basic tasks. I will illustrate these challenges with examples from post-training and evaluation, focusing on core capabilities such as instruction following, factuality, and multilinguality. Addressing these challenges is crucial to realizing the full potential of AI and ensuring its responsible development.

Bio: Prof. Dr. Mario Fritz is a faculty at the CISPA Helmholtz Center for Information Security, an honorary professor at Saarland University, and a fellow of the European Laboratory for Learning and Intelligent Systems (ELLIS). Until 2018, he led a research group at the Max Planck Institute for Computer Science. Previously, he was a PostDoc at the International Computer Science Institute (ICSI) and UC Berkeley after receiving his PhD from TU Darmstadt and studying computer science at FAU Erlangen-Nuremberg. His research focuses on trustworthy artificial intelligence, especially at the intersection of information security and machine learning. He is Associate Editor of the journal “IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI),” coordinates the Helmholtz project “Trustworthy Federated Data Analytics,” and has published over 100 scientific articles – 80 of them in top conferences and journals.

Bio: Roberto Navigli is a full professor of Natural Language Processing in the Department of Computer, Control and Management Engineering at the Sapienza University of Rome, the head of the Sapienza NLP Group, and a Fellow of AAAI, ACL, EurAI and ELLIS. In 2007 he received a Ph.D. in Computer Science from “La Sapienza” (recipient of the Marco Cadoli 2007 AI*IA prize for the Best Ph.D. Thesis in AI). In 2013 he received the Marco Somalvico AI*IA prize, awarded to the best young Italian researcher in AI, who has provided a significant, highly-innovative personal contribution in AI.

He is the creator of BabelNet, the largest multilingual encyclopedic computational dictionary, winner of the 2017 prominent paper award of the Artificial Intelligence Journal and the META prize, and covered by the Guardian and Time Magazine. Also winner of the 2023 prominent paper award of the AI Journal for NASARI and outstanding/best paper awards at NAACL 2021, ACL 2022, 2023 and 2024.

He has received two prestigious grants from the European Research Council (ERC): MOUSSE, a 1.5M euro Consolidator Grant (2017-2022), and MultiJEDI, a 1.3M euro Starting Grant. These grants have been highlighted (out of 10,000) among the 15 projects through which the ERC transformed science.

He is a member of the steering committee of the CINI National Laboratory on Artificial Intelligence and Intelligent Systems.

Abstract: The ability to interpret word meanings in context is a core yet underexplored challenge for Large Language Models (LLMs). While these models demonstrate remarkable linguistic fluency, the extent to which they genuinely grasp word semantics remains an open question. In this talk, we investigate the disambiguation capabilities of state-of-the-art instruction-tuned LLMs, benchmarking their performance against specialized systems designed for Word Sense Disambiguation (WSD). We also examine lexical ambiguity as a persistent challenge in Machine Translation (MT), particularly when dealing with rare or context-dependent word senses. Through an in-depth error analysis of both disambiguation and translation tasks, we reveal systematic weaknesses in LLMs, shedding light on the fundamental challenges they face in semantic interpretation. Furthermore, we show the limitations of standard evaluation metrics in capturing disambiguation performance, reinforcing the need for more targeted evaluation frameworks. By presenting dedicated testbeds, we introduce more effective ways to assess lexical understanding both within and across languages. With this talk we highlight the gap between the impressive fluency of LLMs and their actual semantic comprehension, raising important questions about their reliability in critical applications.