Keynote Talks

Benno Stein Bauhaus-Universität Weimar, Germany

Benno Stein is chair of the Web-Technology and Information Systems Group at the Bauhaus-Universität Weimar. His research focuses on modeling and solving data- and knowledge-intensive information processing tasks. Common ground of his research are the principles and methods of symbolic Artificial Intelligence. Benno Stein has developed theories, algorithms, and tools for information retrieval, data mining, computational linguistics, knowledge processing, as well as for engineering design and simulation (patents granted). For several achievements of his research he has been awarded with scientific and commercial prizes.
Professional background: Study at the University of Karlsruhe (1984-1989). Dissertation (1995) and Habilitation (2002) in computer science at the University of Paderborn. Appointment as full professor for Web Technology and Information Systems at the Bauhaus-Universität Weimar (2005). Research stays at IBM (Germany) and the International Computer Science Institute (Berkeley, US). Benno Stein is an initiator and co-chair of both PAN and Touché, two series of scientific events and shared tasks on digital text forensics and argument retrieval. He is cofounder and spokesperson of the Digital Bauhaus Lab Weimar, a visionary and interdisciplinary research centre for Computer Science, Arts, and Engineering. Not least, he is a cofounder (1996) and scientific director of the Art Systems Software Ltd, a world leading company for simulation technology in fluidic engineering.

Title: [Perceived] Limits in Information Retrieval

Abstract: Can information retrieval systems satisfy the desire for unbiased and unframed information? Moreover, should information retrieval systems satisfy the desire for unbiased and unframed information? And finally, do users of information retrieval systems have such a desire in the first place? We will start with these questions to discuss and shed light on recent and related developments in information retrieval, such as conversational search, argumentative search, direct answers, information quality labels, and ideas to quantify bias.

Rita Cucchiara Università degli Studi di Modena e Reggio Emilia, Italy

Rita Cucchiara has been, since 2005, Full Professor of Computer Science and Engineering at the University of Modena and Reggio Emilia. She is the Director of the interdepartmental Artificial Intelligence Research and Innovation Center (AIRI) and Scientific Responsible of AImageLab, a Research Laboratory in Computer Vision, Machine/Deep Learning, Artificial Intelligence and Multimedia of the Department of Engineering "Enzo Ferrari", and Scientific Advisor of the AI Academy of Modena, an initiative co-funded by the Emilia Romagna Region for research in Artificial Intelligence and technology transfer in local industries. Her research field deals mainly with Artificial Vision and Deep Learning, with more than 350 publications on the subject. Rita Cucchiara holds a Degree in Electronic Engineering (1989) and a PhD in Computer Engineering (1993), both received at the University of Bologna.

Title: A journey into image captioning research

Abstract: Image captioning is a fashionable research field in AI: as in Neuroscience, only recently the link between human vision and language generation has been clarified. Also in Deep Learning recent architectures, both recurrent and self-attentive, have shown their capabilities in multimodal understanding and generation. Image captioning is indeed the task of describing the visual content of an image in natural language, employing a visual understanding system and a language model capable of generating meaningful and syntactically correct sentences. Research is explored in different directions: on the one hand, it is going toward very large-scale foundation models, with large-scale parameters and very large web-based supervised datasets; on the other hand, new paradigms mixing generative self-attentive architectures and information retrieval explore new problems as for instance zero-shot learning, long-tail concept description enriched by proper names of persons, places and events. This talk will present a brief overview of recent research results and some architecture models developed at AImagelab, University of Modena and Reggio Emilia, for controllable captioning, universal captioning and retrieval-augmented captioning and discusses possibly applications in e-commerce, robotics and web mining, supported by Italian, European and PNRR cofounded projects.