Distinguished Speaker Series
In the ELLIS Distinguished Speaker Series ELLIS Alicante invites top AI researchers from around the world to present their work to the team of ELLIS Alicante. In collaboration with the University of Alicante, the event is open to any student and researcher of the university.
Talks
-
On the origins of theory of mind and misbeliefs – a game theoretical perspective
Abstract: Within AI research, proposals have been launched that argue for the creation of a new science of cooperative AI to address the problems that may arise from the diverse ecologies of AI systems that interact in complex ways with other AI and humans. One element of this new science would be the realization of AI that understands the motivations and ambitions of the other AI and humans in the mix. Such an understanding requires AI to have a Theory of Mind (ToM), i.e., they need to recognize that the other entities have cognitive and emotional states and use that information to make decisions. Many have argued that current LLM-based AI already have that capacity, yet there is no consensus that this is due to actual social reasoning.
In this seminar, Prof. Tom Lenaerts will not address whether the former is true or not, but will present a minimal model that aims to explain when ToM agents are favoured in an ecosystem of interacting agents and what the consequences are. To this end, the authors developed an Evolutionary Game Theoretical model in which a finite population of individuals use strategies that incorporate (or not) a ToM, modelled using level-k recursive reasoning, to infer a best response to the anticipated behaviour of others within the context of the centipede game. They calibrated their results to the existing experimental knowledge on how humans behave in this game. They find that strategies incorporating a ToM evolve and prevail under natural selection, provided the agents make cognitive errors and a temptation for higher future gains is in place. They found furthermore that such non-deterministic reasoning co-evolves with an optimism bias, favouring the selection of a new equilibrium configuration in the centipede game, which was not anticipated to date. This work reveals not only a unique perspective on the evolution of bounded rationality but also a co-evolutionary link between the evolution of ToM and the emergence of misbeliefs.
Short bio: Tom Lenaerts is full professor in the Computer Science department at the Université Libre de Bruxelles (ULB), where he is co-heading the Machine Learning Group, focussing on AI, Game Theoretical modelling and Computational Biology. He holds a partial affiliation as research professor with the Artificial Intelligence Lab of the Vrije Universiteit Brussel and is affiliated researcher at the Center for Human-Compatible AI of UC Berkeley. He was board member, vice-chair and finally chair of the Benelux Association for Artificial Intelligence between 2016 and 2024 and Director of the Interuniversity Institute of Bioinformatics in Brussels between 2017 and 2021. He currently is the Academic Director of FARI, the Brussels AI for Common Good institute and national contact point for the CAIRNE hub in Brussels. He has been publishing in a variety of interdisciplinary domains on AI and Machine Learning, involving topics related to optimization, multi-agent systems, collective intelligence, evolutionary game theory, computational biology and bioinformatics.
Presenter: Tom Lenaerts (Université Libre de Bruxelles)
Date: 10 March 2026 at 12:00 CET
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Evaluating LLM Personas as Proxies for Human Populations
Abstract: This talk presents ongoing work on whether and how LLMs can be guided through personas to simulate human perspectives in support of policy analysis. Recent experiments, including our own, show that persona prompting allows models to reproduce average response patterns. However, these experiments also expose significant limitations: LLMs tend to flatten demographic nuance, produce exaggerated archetypes rather than authentic variation, and systematically collapse the diversity of viewpoints found in real human populations. Results from this project are presented alongside findings from related literature, with particular attention to how these "Artificial Hivemind" behaviors emerge from current training and alignment paradigms. The talk concludes by presenting ongoing research aimed at better aligning LLM outputs with observed human response distributions.
Short bio: Axel Abels is a Postdoctoral Researcher at the Université Libre de Bruxelles' Machine Learning Group. His research focuses on collective intelligence, particularly on methods for improving the wisdom of crowds. His recent work explores hybrid collective intelligence systems that combine human groups and large language models, leading to current research on the capabilities and limits of LLMs in representing human diversity when used as proxies for human populations.
Presenter: Axel Abels (Université Libre de Bruxelles')
Date: 11 February 2026 at 11:00 CET
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
-
NLP applications at UPV to fight against disinformation and sexism
Abstract: Social media offer a fast and easy way for the propagation of disinformation, from fake news to conspiracy narratives. In this talk I will describe the attempt by malicious actors (pseudo-journalists, influencers , far-right political actors and supporters) to trigger negative emotions among the population posting fake news and conspiracy theories in X and TikTok videos on the day of the DANA flood in Valencia at the end of last October, and the days after. Moreover, I will comment on conspiracy theories, complex narratives that attempt to explain the ultimate causes of significant events as cover plots orchestrated by secret, powerful, and malicious groups. Existing AI approaches do not distinguish between critical and conspiratorial thinking, whereas this distinction has important implications for automatic content moderation: without differentiating between critical and conspiratorial thinking, there is a high risk of pushing people toward conspiracy communities. Last, I will briefly mention the challenge on sexism identification in tweets, memes and TikTok videos that we have been co-organising since 2021
Short bio: Paolo Rosso (http://personales.upv.es/prosso/) is Full Professor of Computer Science at the Universitat Politècnica de València, Spain. His current research interests fall mainly in the area of detection of harmful information in social media, both disinformation (fake news and conspiracy theories) and hate speech (with women, immigrants, and LGBTI as targeted groups, and when conveyed in a subtle way via irony, sacarsm or hurtful humour e.g with sexist memes and videos). Recently, he has been the PI of two projects on harmful information detection: XAI-DisInfodemics on eXplainable AI for disinformation and conspiracy detection during infodemics, and FAKEnHATE-PoC: FAKE news and HATE speech, both funded by the Spanish Ministry of Science, Innovation and Universities, and by European Union NextGenerationEU/PRTR. At the moment he is the PI of the FairTransNLP Spanish project on Fairness and Transparency for equitable NLP applications in social media, and a member of the CHIST-ERA MARTINI project on Malicious Actors pRofiling and deTection In online social Network through artificial Intelligence, and the IBERIFIER-plus hub on disinformation detection in Spain and Portugal of the European Digital Media Observatory. He collaborated with the Spanish National Security Department and with the Science and Technology Office (Oficina-C) of the Spanish Congress of Deputies in topics related to disinformation campaigns and AI, and with the Spanish Observatory on racism and xenophobia (OBERAXE).
Presenter: Prof Paolo Rosso (Universitat Politècnica de València, Spain)
Date: 12 June 2025 at 12:30 CEST
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Argumentation: The Missing Piece for Natural Language Reasoning in LLMs
Abstract: Since the release of the latest versions of popular LLMs such as o3, Gemini 2.0 Flash Thinking, or DeepSeekV3-R1, the race for Artificial General Intelligence (AGI) has put its focus on the “reasoning” capabilities of these models. Advances in the “reasoning” capabilities of LLMs are measured and quantified with the use of benchmarks including a wide set of tasks and challenges that address reasoning from different perspectives (e.g., commonsense, mathematical, or multi-hop reasoning). It is interesting, however, to see that none of these tasks involve argumentation, the natural form of reasoning in natural language. In this talk, Ramon will briefly introduce the nature of the “reasoning” capabilities of LLMs, covering some of the tasks included in the benchmarks. Then, fundamental concepts of argumentative reasoning will be presented, pointing out the main differences between reasoning in natural language with arguments, and the tasks used to benchmark reasoning. Finally, the talk will end with a discussion on the most important open challenges that will push forward the boundaries of reasoning in LLMs.
Short bio: Ramon Ruiz-Dolz is a Lecturer in Computing at the University of Dundee (UK). His main areas of research are Computational Argumentation and Natural Language Processing. His current research focuses on the analysis of the natural language reasoning capabilities of LLMs, and the integration of concepts from argumentation theory into NLP algorithms to improve their explainability and to create tools for developing critical thinking skills. Ramon obtained his PhD in Computer Science from the Universitat Politècnica de València (Spain).
Presenter: Ramon Ruiz-Dolz (University of Dundee (UK))
Date: 4 June 2025 at 11:30 CEST
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Cybersecurity in the Age of AI
Abstract: This talk explores the intersection of cybersecurity, federated learning (FL), and large language models (LLMs) as foundational technologies for privacy-preserving and resilient security solutions. Topics include the application of FL for cyberattack detection in distributed and non-IID environments, and the design of robust aggregation mechanisms against data/model poisoning attacks. Additionally, the talk addresses the use of LLMs in cybersecurity tasks such as classification of cyber threat intelligence (CTI) reports and automated fuzzing for vulnerability discovery, highlighting their potential to enhance both data diversity and detection accuracy.
Short bio: José Luis Hernández-Ramos is an Associate Professor in Cybersecurity at the University of Murcia, whose work has made significant contributions to research, policy, and standardization in cybersecurity. He was awarded a prestigious Marie Skłodowska-Curie Fellowship and worked for five years at the European Commission (DG JRC), where he contributed to key initiatives such as the EU Cybersecurity Taxonomy and the Cybersecurity Atlas. He has participated in several EU research and co-authored over 60 scientific publications. He has received multiple awards, including a BBVA Leonardo Grant, and has served as an expert evaluator for various national and international research agencies.
Presenter: Prof José Luis Hernández-Ramos (University of Murcia, Spain)
Date: 4 June 2025 at 10:30 CEST
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
From Biology to Machine Learning and vice versa
Abstract: In this talk, we will revisit the intertwined history of neuroscience and machine learning, highlighting the many connections between the two fields — especially when dealing with natural images and the concept of invariance. Humans are naturally invariant to a wide range of transformations such as translation, scaling, and contrast changes. Understanding how our brains achieve this robustness could lead to the development of much more efficient artificial neural networks. We will explore how biologically inspired models, particularly those mimicking the structure and function of the visual cortex, have influenced modern ML architectures . Finally, we will present a specific use case where following a “human-inspired” approach led to improved performance of neural networks in blurred-image classification. Using a method based on entropy maximization, we show how pre-processing strategies inspired by biological systems can significantly narrow the gap between machine and human perception. This suggests that drawing deeper from biology not only helps us understand our brain better but also paves the way for more robust and interpretable machine learning systems.
Short bio: Prof. Marina Martinez-Garcia received the B.S. degree in mathematics from the Polytechnic University of Catalonia (UPC), Spain, in 2009, and the Ph.D. degree in computer science from Pompeu Fabra University (UPF) in 2014. She is a Professor in the mathematics Department of Universitat Jaume I, Castelló Spain. She is also a member of the Institut de Matemàtiques de Castello (IMAC) and the Instituto de Estudios Feministas Purificación Escribano. Her current research interests include machine learning and deep learning with applications in neuroscience.
This talk is part of the 3rd Distinguished Lectures Mini-Workshop.
Presenter: Prof. Marina Martinez-Garcia (Universitat Jaume I, Castelló Spain)
Date: 15 May 2025 at 12:15 CEST
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
The Impact of Ranking Interventions and Task-Level Factors in Recruitment Interfaces for Shortlisting
Abstract: Personnel recruitment is increasingly mediated by automated mechanisms to select and rank candidates. We present a series of user studies on candidate shortlisting that seek to elucidate the way in which selecting candidates from a ranked list is affected by task-level factors, individual representations of candidates, and the ordering in which candidates are displayed. In a nutshell, demographic balance in ranking visibility does not entail demographic balance in outcomes. These findings have implications for the design of algorithms, interfaces, and initiatives for diversity in employment. This talk corresponds to joint work with Alessandro Fabbris, Clara Rus, Asia Biega, Anna Gatzioura, and Jorge Saldivar within Horizon Europe project FINDHR.
Short bio: Carlos Castillo (they/them) is an ICREA Research Professor at Universitat Pompeu Fabra in Barcelona, where they lead the Web Science and Social Computing research group. They are a web miner with a background in information retrieval and have been influential in the areas of crisis informatics, web content quality and credibility, and adversarial web search. They are a prolific, highly cited researcher who has co-authored over 110 publications in top-tier international conferences and journals, receiving two test-of-time awards, five best paper awards, and two best student paper awards. Their works include a book on Big Crisis Data, as well as monographs on Information and Influence Propagation, and Adversarial Web Search.
This talk is part of the 3rd Distinguished Lectures Mini-Workshop.
Presenter: Prof Carlos Castillo (University Pompeu Fabra, Spain)
Date: 15 May 2025 at 11:30 CEST
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Minds in Machines? On the Psychology and Deception Capabilities of Large Language Models
Abstract: Large language models (LLMs) are currently at the forefront of intertwining AI systems with human communication and everyday life. Therefore, it is of great importance to thoroughly assess and scrutinize their capabilities. Due to the increasingly complex and novel behaviors of today's LLMs, this can be done by treating them as participants in psychological experiments originally designed to test humans. The talk will outline how psychology can inform behavioral tests for LLMs, and how such tests can discover emergent abilities in LLMs. As a deep dive, the talk will focus on the phenomenon of deception in LLMs - an ability that carries significant implications for alignment and safety.
Short bio: Dr. Thilo Hagendorff is an expert in AI safety, AI ethics, machine behavior in generative models, as well as the intersection of machine learning and psychology. He is working as a Research Group Leader at the University of Stuttgart. Previously, he worked for the Cluster of Excellence “Machine Learning” at the University of Tuebingen. He was a visiting scholar at Stanford University as well as UC San Diego. As a lecturer, he teaches at the Hasso Plattner Institute in Potsdam, among others.
Presenter: Dr. Thilo Hagendorff (University of Stuttgart, Germany)
Date: 9 May 2025 at 10:30 CEST
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Urban Safety Perception Through the Lens of Large Multimodal Models
Abstract: Understanding how urban environments are perceived in terms of safety is crucial for urban planning and policymaking. Traditional methods like surveys are limited by high cost, required time, and scalability issues. To overcome these challenges, this study introduces Large Multimodal Models (LMMs), specifically Llava 1.6 7B, as a novel approach to assess safety perceptions of urban spaces using street-view images. In addition, the research investigated how this task is affected by different socio-demographic perspectives, simulated by the model through Persona-based prompts. The model achieved an average F1-score of 59.21% in classifying urban scenarios as safe or unsafe. Incorporating Persona-based prompts revealed significant variations in safety perceptions across the socio-demographic groups of age, gender, and nationality. Elder and female Personas consistently perceive higher levels of unsafety than younger or male Personas. Similarly, nationality-specific differences were evident in the proportion of unsafe classifications ranging from 19.71% in Singapore to 40.15% in Botswana. Notably, the model’s default configuration aligned most closely with a middle-aged, male Persona. Interestingly, model's decisions are based on visual features like isolation and lack of maintenance that are in line with multiple urban social theories like Eyes on the Street and Defensible Space. Our findings highlight the potential of LMMs as a scalable and cost-effective alternative to traditional methods for urban safety perceptions. While the sensitivity of these models to socio-demographic factors underscores the need for thoughtful deployment, their ability to provide nuanced perspectives makes them a promising tool for AI-driven urban planning.
Short bio: Dr. Massimiliano Luca is a senior researcher at the Mobile and Social Computing Lab - Center for Augmented Intelligence - Fondazione Bruno Kessler, Trento, and a member of the Complex System Society Council. In November 2023, he obtained his PhD at the Free University of Bolzano/Bozen and, during his PhD he had the opportunity to be a Visitor Researcher at MIT, Boston and at the Spanish National Research Council (CSIC). He authored more than 20 research papers focused on artificial intelligence applied to mobility, transportation and city science. Such publications received more than 5 awards from different conferences and journals. Previously, Massimiliano was the Chief Scientific Officer at Pulse.io, a London-based start-up that employs artificial intelligence in emotional marketing where he was responsible for data science, web-services implementation, cloud computing, deep learning and generative AI assets.
This talk is part of the ELLIS / FBK-MobS Workshop.
Presenter: Dr. Massimiliano Luca (Bruno Kessler Foundation, Trento, Italy)
Date: 9 April 2025 at 12:15 CEST
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Trustworthy AI and Compositionality in Multimodal Language Models
Abstract: This presentation will give an overview of two different perspectives on trustworthy AI; one technical and one sociotechnical view. The technical view focuses on visual reasoning task experiments with multimodal LLMs. These experiments show how LLMs fail to compositional generalise and their lack of consistency in problem solving. The experiments are based on a novel benchmark with synthetic data from the CLEVR domain. The sociotechnical view is based on critical analysis of Reinforcement Learning from Human Feedback as a method for AI alignment. This perspective will show all the ways in which RLHF is insufficient as a method, and the fundamental issues with the concept of alignment.
Short bio: Adam Dahlgren Lindström (PhD 2024) is a postdoctoral research fellow in the Computing Science department at Umeå University, Sweden, where he is a part of the Responsible AI research group and AI policy lab. His thesis on Learning, Reasoning, and Compositional Generalisation in Multimodal Language Models, investigated limitations in vision-language models and how to better measure phenomena related to reasoning tasks and compositionality. During his PhD, he was affiliated with Swedens largest research program, WASP, and sat as one of the conference chairs for the int. conf. on Hybrid Human-AI Intelligence (HHAI) 2024. He is currently funded by the ELIAS project, where he is part of the tasks on trustworthy AI and hybrid AI-methods. In his current research, he is working on the benchmarking of multimodal models, e.g. for reasoning capabilities and ontological knowledge, and on methodologies for hybrid human-AI collaboration and sustainable deployment of AI technologies.
Presenter: Adam Dahlgren (Umeå University, Sweden)
Date: 26 March 2025 at 10:00 CET
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Enhancing Collective Intelligence through Learned Aggregation
Abstract: Human decision-making is inherently flawed by individual and social biases that distort our perception of truth, often leading to sub-optimal group decisions. We explore an alternative to consensus decision-making: aggregating independent judgments to minimize the effects of biases such as groupthink and herding. The key hypothesis is that by carefully optimizing the collectivization of knowledge, it will be substantially harder for individuals to impose their biases on the final decision. The core of our work therefore involves the development and analysis of algorithms designed to effectively aggregate diverse sources of expertise. We focus on transparent aggregation methods that use online machine/reinforcement learning to take into account the nuances of individual expertise and the impact of biases, aiming to filter out noise and enhance the reliability of collective decision-making under uncertainty. Our findings demonstrate a marked improvement in decision-making accuracy and a reduction in bias, underscoring the potential of technology-assisted methods in fostering collective intelligence. We also show how collectives benefit from the inclusion of AI participants who complement human strengths, leading hybrid groups to improved outcomes when compared to either purely human or artificial groups.
Short bio: Axel Abels is currently a post-doctoral researcher affiliated to the Université Libre de Bruxelles' Machine Learning Group and the Vrije Universiteit Brussel's AI Lab. He studied Computer Science at the Université Libre de Bruxelles where his master's thesis extended the applicability of Deep Reinforcement Learning to Multi-Objective problems. His PhD research focused on Collective Decision-Making (CDM), and more specifically how machine/reinforcement learning can be applied to extract the wisdom of the crowd. His current research builds on his PhD, further exploring fairness in CDM as well as the benefits of human-AI hybrid crowds.
This talk is part of the ELLIS / ULB-MLG Workshop.
Presenter: Axel Abels (Université Libre de Bruxelles)
Date: 28 February 2025 at 12:00 CET
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
On defaults and delegation in decision-making – Introducing the labs and some of the ongoing human-AI interaction projects
Abstract: After an introduction of the Brussels’ AI ecosystem and some of the research lines currently explored in the team, this presentation will zoom in on two projects recently completed. The first asked the question how people deal with defaults in decision-making tasks. We often ignore the default settings just to be able to use some platform, but this may have consequences. I will show what to expect from pro-social versus pro-self defaults in a resource extraction problem (e.g. energy sharing). The second project explored the question whether delegating decision-making to an AI leads to better societal outcomes than when people act themselves. Outsourcing to an AI may be better given the potential for a conflict of interest between a principal and human agent. In the context of a collective risk dilemma (e.g. climate action problem), I will show how switching from a human agent to an artificial one affects the decision-making process. Both projects are part of our Social AI program where we aim to, on the one hand, understand the impact of using AI systems in decision-making processes and, on the other hand, design AI systems to achieve beneficial and fair outcomes, both for the user and society.
Short bio: Tom Lenaerts (PhD 2003) is Professor in the Computer Science department at the Université Libre de Bruxelles (ULB), where he is co-heading the Machine Learning Group (MLG). He holds also a partial affiliation as research professor with the Artificial Intelligence Lab of the Vrije Universiteit Brussel and is affiliated researcher at the Center for Human-Compatible AI of UC Berkeley. He was board member, vice-chair and finally chair of the Benelux Association for Artificial Intelligence between 2016 and 2024. He currently is one of the academic directors of FARI, the Brussels AI for Common Good institute, AI expert in the Global Partnership on Artificial Intelligence and national contact point for the CAIRNE hub in Brussels. He has been publishing in a variety of interdisciplinary domains on AI and Machine Learning, involving topics related to optimization, multi-agent systems, collective intelligence, evolutionary game theory, AI governance, computational biology and bioinformatics.
This talk is part of the ELLIS / ULB-MLG Workshop.
Presenter: Tom Lenaerts (Université Libre de Bruxelles)
Date: 28 February 2025 at 11:30 CET
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
AnomalousNet: A Hybrid Approach with Attention U-Nets and Change Point Detection for Accurate Characterization of Anomalous Diffusion in Video Data
Abstract: Anomalous diffusion occurs in various systems, including protein transport within cells, animal movement in complex habitats, pollutant dispersion in groundwater, and nanoparticle motion in synthetic materials. Accurately estimating the anomalous diffusion exponent and diffusion coefficient from particle trajectories provide deeper insights into the system’s underlying dynamics, facilitating the identification of particle behaviors and detecting changes in diffusion states. However, analyzing short and noisy video data, which often yield heterogeneous and incomplete trajectories, poses a significant challenge for traditional statistical approaches. To address these issues, we introduce a data-driven method that integrates particle tracking, an Attention U-Net architecture, and change-point detection. Our methodology demonstrated strong performance in the 2nd Anomalous Diffusion (AnDi) Challenge, ranking within the top two submissions for the single-trajectory tasks in the video-based track and among the top six for the ensemble task in the single-trajectory track.
Short bio: J. Alberto Conejero is full professor in the Department of Applied Mathematics of Universitat Politècnica de València (UPV) since 2020. He is also Vice-rector of Students and Entrepreneurship at UPV. Before, he has also been Deputy Secretary General and Director of the Academic Academic Performance and Curricular Assessment Area. He has published more than 100 articles in indexed journals on mathematics and its applications. In 2014 he received the Teaching Excellence Award from the Social Council of UPV, and in in 2021 was co-leader of the winning team of the Pandemic Response Challenge of the XPRIZE Foundation.
This talk is part of the 2nd Distinguishes Lectures Mini-Workshop.
Presenter: J. Alberto Conejero (Universitat Politècnica de València, València)
Date: 20 February 2025 at 12:15 CET
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
The role of artificial intelligence in achieving the Sustainable Development Goals
Abstract: The emergence of artificial intelligence (AI) and its progressively wider impact on many sectors requires an assessment of its effect on the achievement of the Sustainable Development Goals. Using a consensus-based expert elicitation process, we find that AI can enable the accomplishment of 134 targets across all the goals, but it may also inhibit 59 targets. However, current research foci overlook important aspects. The fast development of AI needs to be supported by the necessary regulatory insight and oversight for AI-based technologies to enable sustainable development. Failure to do so could result in gaps in transparency, safety, and ethical standards.
Short bio: Dr. Ricardo Vinuesa is an Associate Professor at the Department of Engineering Mechanics, KTH Royal Institute of Technology in Stockholm.
He is also Lead Faculty at the KTH Climate Action Centre. He studied Mechanical Engineering at the Polytechnic University of Valencia (Spain), and he received his PhD in Mechanical and Aerospace Engineering from the Illinois Institute of Technology in Chicago. His research combines numerical simulations and data-driven methods to understand, control and predict complex wall-bounded turbulent flows, such as the boundary layers developing around wings and urban environments. Dr. Vinuesa has received, among others, an ERC Consolidator Grant, the TSFP Kasagi Award, the MST Emerging Leaders Award, the Goran Gustafsson Award for Young Researchers, the IIT Outstanding Young Alumnus Award, the SARES Young Researcher Award and he leads several large Horizon Europe projects. He is also a member of the Young Academy of Science of Spain.This talk is part of the 2nd Distinguishes Lectures Mini-Workshop.
Presenter: Dr. Ricardo Vinuesa (KTH Royal Institute of Technology, Stockholm)
Date: 20 February 2025 at 11:30 CET
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Human-Centered AI: Designing for Urban AI Interactions and Literacy
Abstract: Applications of Artificial Intelligence are entering all areas of society, including urban environments. While research and development of AI technologies have taken major leaps, understanding the human perspective of AI is still in its early stages. In my talk I will focus on how AI’s design qualities, such as proactivity, capability to learn, and embodiment, can be used to advance citizens’ experiences and sustainable living. I will present two cases of design research for urban AI interactions: Intelligent agents for residential community interactions in Nordic Superblocks, and advancing AI transparency and literacy in public urban spaces. In my talk I will also suggest future directions for research and practice of human-centred urban AI.
Short bio: Kaisa Väänänen is a full professor of Human-Technology Interaction in Tampere University, Finland. Kaisa has over 25 years of experience of interdisciplinary research both in industry and academia. Kaisa is passionate about understanding how user needs and experiences can be supported by novel technologies. Currently, she focuses on Human-Centered AI design research and how AI-powered solutions can advance sustainability. Kaisa was the general co-chair of ACM CHI’23 conference, and in 2025, she is co-organising Intelligent User Interfaces, IUI’25. In 2022, Kaisa was selected as an ACM Distinguished Member for her long-standing contribution to the field of computing. https://research.tuni.fi/kaisa-vaananen/
Presenter: Prof. Kaisa Väänänen, Human-Technology Interaction, Tampere University, (Tampere, Finland)
Date: 3 October 2024 at 11:30 CEST
Location: Oficinas ELLIS Alicante, Muelle Pte., 5 – Edificio A, Alicante 03001, Alicante ES
Online: Meeting link
-
Breaking Bad Bias: Gender Stereotypes in Generative Models
Abstract: Generating images from textual descriptions requires the generative model to make implicit assumptions about the output scene that are not explicitly instructed in the input prompt. These assumptions can reinforce unfair stereotypes related to gender, race, or socioeconomic status. However, measuring and quantifying these social biases in generated images is a big challenge. In this talk, we will explore methods for measuring gender bias in text-to-image models, particularly Stable Diffusion, and discuss how the generated images, when used to train future computer vision models, affect bias in downstream tasks.
Short bio: Noa Garcia is an Associate Professor at the Institute for Advanced Co-Creation Studies, Osaka University, Japan. Originally from Barcelona, she moved to Japan in 2018, first as a postdoctoral researcher and then as a specially-appointed assistant professor at the Institute for Datability Science. She completed her Ph.D in multimodal retrieval and instance-level recognition at Aston University, United Kingdom, after earning her degree in Telecommunications Engineering from Universitat Politècnica de Catalunya, Barcelona. Her current research interests lie at the intersection of computer vision, natural language processing, fairness, and art. She is an active member of the computer vision community, having co-organized several workshops and international events, and regularly publishes at top conferences such as CVPR, ICCV, ECCV, or NeurIPS.
Presenter: Asst. Prof. Noa García, Computer Vision, Osaka University (Osaka, Japan)
Date: 27 September 2024 at 11:30 CEST
Location: Distrito Digital 5, Muelle Pte., 5 – Edificio D, Alicante 03001, Alicante ES
Online: Meeting link
-
Using AI bias for good: poverty and inequality mitigation
Abstract: The talk is framed in the AI for Good interdisciplinary area of work, which seeks to direct AI research towards the advancement of the United Nations Sustainable Development Goals. In particular, I will be presenting lines of research in AI-enabled tools that aim to open new paths for poverty reduction by acting on social discrimination. While bias mitigation in AI has generated an important body of publications, we argue that online bias can be useful to identify and measure shared beliefs that influence social policy making. In this talk, I will outline research directions that aim to generate a global index on discrimination against the poor (through NLP and LLMs), and to optimize poverty-mitigation policies via AI simulations (Agent-Based Modeling).
Short bio: Georgina Curto is an Assistant Research Professor at the Lucy Family Institute for Data and Society, University of Notre Dame. She chairs the IJCAI Symposia in the Global South and co-chairs the AI & Social Good Special Track at the International Joint Conference on Artificial Intelligence (IJCAI'23). Focusing on issues of poverty mitigation, fairness and inclusion, she works on the design and use of AI socio-technical systems (including NLP, Agent-Based Modeling, Social Networks, Machine Learning and GenAI) to advance interdisciplinary research towards the achievement of the UN Sustainable Development Goals (SDGs).
This talk is part of the Distinguishes Lectures Mini-Workshop.
Presenter: Asst. Res. Prof. Giorgina Curto Rex, Lucy Family Institute for Data & Society, University of Notre Dame, (Notre Dame, United States)
Date: 11 July 2024 at 12:00 CEST
Location: Distrito Digital 5, Muelle Pte., 5 – Edificio D, Alicante 03001, Alicante ES
Online: Meeting link
-
The search for personality markers in text: Challenges for contemporary psychological science
Abstract: The lecture will introduce the field of psychology of language and word use, both in terms of its goals, methods, and results. The primary focus will be on the method of computational psychological-linguistic text analysis, which has dominated the field for the last three decades. Attention will be paid to the elaboration of main psycho-linguistic projects, which examined relations between the use of linguistic categories in written/oral texts and personality characteristics of the author. Different levels of the text will be discussed, especially linguistic morphology, syntax and stylistics. We focus on comparison of the studies, summarize available results, provides their interpretations in a cross-linguistic and cross-situational perspective. We conclude the lecture by presenting key challenges for contemporary (not only) psychological science.
Short bio: Dalibor Kučera is a teacher and researcher in the field of general, social and educational psychology. His long-term professional focus is research in the field of methods based on communication analysis and their application in psychology. The key topic he has been working on since 2012 is the field of psychology of language use. In 2016-2018, he was the principal investigator of a three-year CPACT research project, “Computational Psycholinguistic Analysis of Czech Text”, supported by Czech Science Foundation (16-19087S). In 2020, he was awarded a J. W. Fulbright-Masaryk Senior Fulbright Scholarship with the project “Personality Processes and Oral Communication” (2020-28-11) at the University of Arizona (Tucson). He holds a Ph.D. degree in general psychology and a doctorate in psychology at the Faculty of Arts of Masaryk University, and an associate professorship in psychology at the Faculty of Arts of Charles University. Dalibor Kučera is the author of several professional publications, including the books “Personality Markers in Text”, devoted to the application of quantitative psychological-linguistic analysis of text in the description of personality and “Modern Psychology: The Main Fields and Topics of Contemporary Psychological Science”
This talk is part of the Distinguishes Lectures Mini-Workshop.
Presenter: Assoc. Prof. Dalibor Kučera, General, Social and Educational Psychology, University of South Bohemia, (České Budějovice, Czech Republic)
Date: 11 July 2024 at 11:15 CEST
Location: Distrito Digital 5, Muelle Pte., 5 – Edificio D, Alicante 03001, Alicante ES
Online: Meeting link
-
Exploring NLP Advancements: Addressing Biases and Human Values in Large Language Models
Abstract: This talk will introduce the Natural Language and Text Processing Lab (NLTP) at Utrecht University (https://nlp.sites.uu.nl/), highlighting its focus on applications of NLP and large language models. The lab conducts cutting-edge research, exploring various domains where NLP can be transformative. An overview of selected ongoing projects will be given, showing how advanced NLP techniques are used to solve real-world problems. The second part of the talk will focus on a recent study investigating the sensitivity of large language models to human values and biases. These models often reflect societal and cultural biases present in their training data. While previous research has shown that monolingual models can exhibit moral biases, there is a gap in understanding how these biases manifest in different cultural contexts. This study investigates the extent to which monolingual and multilingual models encode knowledge of moral norms from different countries. Explainable NLP tools are used to interpret the inferences made by these models, providing insights into how they incorporate and reflect moral variation across cultures.
Short bio: Dr. Ayoub Bagheri is Associate Professor of NLP and Data Science at the Department of Methods and Statistics at Utrecht University. He leads the NLTP lab, which focuses on NLP and AI methods applied to textual data. His main interests include fundamental research in text mining and NLP, with particular emphasis on areas such as bias and personality detection, explainable AI, large language models, and practical applications of language models in computational social sciences and health research. Dr. Bagheri is a board member of the Utrecht Young Academy (UYA) and the Centre for Unusual Collaborations in the Netherlands. He is one of the general chairs of BNAIC/BeNeLearn 2024 and a regular committee member in various NLP conferences, including ACL, LREC-COLING, ECAI and EMNLP.
This talk is part of the Distinguishes Lectures Mini-Workshop.
Presenter: Assoc. Prof. Ayoub Bagheri, NLP and Data science, Utrecht University (Utrecht, Netherlands)
Date: 11 July 2024 at 10:15 CEST
Location: Distrito Digital 5, Muelle Pte., 5 – Edificio D, Alicante 03001, Alicante ES
Online: Meeting link
-
AI-Driven Personalization to Support Human-AI Collaboration
Abstract: At the Human-AI Interaction group at the University of British Columbia, we investigate how to support Human-AI collaboration via AI artifacts that can understand relevant properties of their users (e.g., states, skills, needs) and personalize the interaction accordingly in a manner that preserves transparency, user control and trust. In this talk, I will illustrate examples of our research in AI-drive personalization spanning areas such as User Adaptive Visualizations, intelligent Tutoring Systems, and Personalized Explainable AI.
Short bio: Dr. Conati is a Professor of Computer Science at the University of British Columbia, Vancouver, Canada. She received a M.Sc. in Computer Science at the University of Milan, as well as a M.Sc. and Ph.D. in Intelligent Systems at the University of Pittsburgh. Cristina has been researching human-centered and AI-driven personalization for over 25 years, with contributions in the areas of Intelligent Tutoring Systems, User Modeling, Affective Computing, Information Visualization and Explainable AI.
Cristina's research has received 10 Best Paper Awards from a variety of venues, as well as the Test of Time Time Award 2022 from the educational data mining society. She is a Fellow of AAAI (Association for the Advancement of AI) and of AAIA ( Asia-Pacific Artificial Intelligence Association ), an ACM Distinguished Member, and an associate editor for UMUAI (Journal od User Modeling and User Adapted Interaction), ACM Transactions on Intelligent Interactive Systems and the Journal of Artificial Intelligence in Education. She served as President of AAAC, (Association for the Advancement of Affective Computing), as well as Program or Conference Chair for several international conferences, including UMAP, ACM IUI, and AI in Education.This talk is part of the Distinguishes Lectures Mini-Workshop.
Presenter: Prof. Cristina Conati, Computer Science, University of British Columbia, (Vancouver, Canada)
Date: 11 July 2024 at 09:30 CEST
Location: Distrito Digital 5, Muelle Pte., 5 – Edificio D, Alicante 03001, Alicante ES
Online: Meeting link
-
AI for social impact
Abstract: How can data science and AI make a positive impact in our society and our planet? Scientific work across disciplines can have methodological, operational and policy impacts. I´ll share projects and lessons learned through my journey in academia, international organizations and as an entrepreneur where data science, complex systems and AI tools have been used in challenges from many different disciplines including basic biology research, personalized medicine, infodemics, human rights, or humanitarian response. I´ll also discuss in detail the work of Spotlab.ai developing and deploying AI for diagnostics and clinical research in applications from neglected tropical diseases to onco- hematological diseases.
Short bio: Scientist, entrepreneur, policy advisor and engineering professor passionate about imagining, building and sharing responsible AI for humanity and the planet. Dr. Miguel Luengo-Oroz is the founder and CEO of Spotlab.ai, an AI platform for clinical research and universal diagnosis. Miguel is the former first Chief Data Scientist at the United Nations and has been the head of the data science teams across the network of UN Global Pulse labs. He has worked in many domains including global public health, infodemics, poverty, food security, refugees & migrants, conflict prevention, human rights, privacy, gender, hate speech, and climate. He is also a Professor at the doctoral school and board member at the Telecommunications School of the Universidad Politécnica de Madrid. Miguel serves on multiple international advisory boards around responsible AI and advises leadership on the impact of macro trends and frontier technologies for their organizations and society. Miguel has been awarded as an Obama Foundation Leader, Ashoka fellow, MIT TR35, EU Responsible Research & Innovation award, FRSA-UK, Red Cross Innovation award and La Caixa fellow. He holds a Ph.D. and MSc.Eng from the Universidad Politécnica de Madrid and a MSc from the Ecole des Hautes Etudes en Sciences Sociales de Paris.
Presenter: Dr. Miguel Luengo, Chief Data Scientist, Spotlab, (Madrid, Spain)
Date: 31 January 2024 at 14:00 CET
Location: Distrito Digital 5, Muelle Pte., 5 – Edificio D, Alicante 03001, Alicante ES
-
Towards vision based Emotion AI
Abstract: Emotions play a key role in human-human interactions and become one key focus in future Artificial Intelligence. There is a growing need to develop emotionally intelligent interfaces, which are able to read the emotions of the users and adapt their operations accordingly. Among the areas of application are human-robot interaction, emotional chatpots, health and medicine, on-line learning, user or customer analysis, and security and safety. This talk will provide an introduction to the emotional interfaces, and overviews our progress in related research. The research topics to be covered include facial (micro)-expression recognition, emotional gestures, remote heart rate measurement from videos and potential applications. Finally, some future challenges are outlined.
Short bio: Guoying Zhao received the Ph.D. degree in computer science from the Chinese Academy of Sciences, Beijing, China, in 2005. She is currently an Academy Professor and full Professor (tenured in 2017) with University of Oulu. She is/was also a visiting professor with Aalto University and Stanford University. She is a member of Academia Europaea, a member of Finnish Academy of Sciences and Letters, Fellow of IEEE, IAPR and ELLIS. She was panel chair for IEEE conference on Automatic Face and Gesture (FG 2023), publicity chair of 22nd Scandinavian Conference on Image Analysis (SCIA 2023), co-program chair for ACM International Conference on Multimodal Interaction (ICMI 2021), co-publicity chair for FG2018, and has served as associate editor for IEEE Trans. on Multimedia, Pattern Recognition, IEEE Trans. on Circuits and Systems for Video Technology, Image and Vision Computing and Frontiers in Psychology Journals. Her current research interests include image and video descriptors, facial-expression and micro-expression recognition, emotional gesture analysis, affective computing, and biometrics. Her research has been reported by Finnish TV programs, newspapers and MIT Technology Review.
Presenter: Prof. Guoying Zhao, Computer Science and Engineering, University of Oulu,(Oulu, Finland)
Date: 23 November 2023 at 11:30 CET
Location: Salón de Grados del Edificio de Óptica, Carretera San Vicente del Raspeig s/n, San Vicente del Raspeig 03690, Alicante ES
Online: Meeting link
-
Towards Unbiased LLMs from the Roots: Exploring Biases in Language Corpora
Abstract: In the rapidly advancing field of Natural Language Processing (NLP), driven by the widespread adoption of Large Language Models (LLMs), biases inherent in these models mirror the broader societal biases present in the textual data they are trained on. One of the typical examples is the gender bias. For instance, LLMs inadvertently perpetuate stereotypes by linking certain professions or characteristics more strongly with a particular gender. This talk will examine the entire pipeline associated with biases in textual data sets (language corpora). We will discuss the most prevalent types of biases, investigate the methods to measure them and suggest techniques for debiasing the data. The talk will present relevant related work and provide useful insights into practical strategies for identifying and mitigating biases in textual data.
Short bio: Erik Derner received his Ph.D. in Robotics and Artificial Intelligence from the Czech Technical University in Prague, Czech Republic, in 2022. Currently, he is an ELLIS postdoctoral researcher at ELLIS Alicante, working on human-centric AI research in the team of Dr. Nuria Oliver. The main objective of his research is to contribute to the development of fair and safe language models, specifically focusing on low-resource languages. His areas of interest comprise human-centric AI, large language models, robotics, computer vision, reinforcement learning, and genetic algorithms. He is a member of the ELLIS network.
Presenter: Dr. Erik Derner
Date: 8 November 2023 at 11:30 CET
Location: Salon de Actos Politecnica IV, Carretera San Vicente del Raspeig s/n, San Vicente del Raspeig 03690, Alicante ES
-
Fast approximate matrix multiplication – theory and practice in AI hardware
Abstract: Multiplying large, dense matrices is a key ingredient of deep networks, and state of the art architectures may spend more than 99% of time and energy consumption on matrix multiplication. Although fast matrix multiplication algorithms in time o(n^3) are known in the theoretical literature, they are impractical, and not used in the AI hardware and software ecosystem today.
In this talk I will survey algorithms and heuristics for faster approximate matrix multiplication, and show the possible impact, as well as limitations, of such approaches on the future of deep learning. The results I will present will cover both academic research, as well as my more recent experience with cutting edge AI industry at deci.ai, where I have discovered a huge gap between the academic research and practical AI.Short bio: Nir Ailon is a Computer Science professor at Technion in Haifa Israel. He completed his PhD in 2006 at Princeton University, and then continued as a postdoc at the Institute for Advanced Study in Princeton. He has served as faculty at Technion since 2011, and also spent time at Google Research, Yahoo! Research, as well as at other industrial research labs. He is a recipient of an ERC Consolidator Grant. His research spans theory of algorithms, mathematical foundations of big data and machine learning. He is currently involved in research on AI acceleration at deci.ai
Presenter: Assoc. Prof. Nir Ailon, Computer Science, Technion Israel Institute of Technology, (Haifa, Israel)
Date: 7 April 2022 at 11:30 CEST
Location: Salon de Actos Politecnica IV, Carretera San Vicente del Raspeig s/n, San Vicente del Raspeig 03690, Alicante ES
-
AI Ethics: a challenging task
Abstract: -
Presenter: Dr. Ricardo Baeza-Yates, Director of Research, Institute for Experiential AI of Northeastern University (San Jose, CA, United States)
Date: 8 November 2021 at 11:30 CET
Location: University of Alicante, Carretera San Vicente del Raspeig s/n, San Vicente del Raspeig 03690, Alicante ES