Category Archives: Highlights

DEI Talks | “Accelerating ML for Science Applications” by Prof. Seda Ogrenci (Northwestern University)

Posted on

The talk “Accelerating ML for Science Applications” will be presented by Prof. Seda Ogrenci (McCormick School of Engineering, Northwestern University) on March the 12th, at 11:00, in room I-105. The session will be moderated by Tiago Carvalho (DEI).

About the Talk:

“Emerging open-source tools and methodologies targeting reconfigurable fabrics hold significant promise for lowering barriers to research, education, and innovation. There are exciting developments in diverse domains where such benefits are demonstrated. This talk will review active domains with needs and applications for real-time ultra low latency ML hardware and how open-source tools need to evolve to provide a multitude of features to enable design of hardware efficient and adaptive ML. As part of this discussion, examples of research directions in adaptive and resilient ML hardware synthesis flows developed in Dr. Ogrenci’s lab will presented.”

About the Speaker:

Seda Ogrenci is a Professor in the Department of Electrical and Computer Engineering (ECE), and in the Department of Computer Science (CS). She is the Director of the Computer Engineering Division of ECE. She has received her PhD degree in Computer Science from the University of California-Los Angeles. She is the co-author of over 140 peer reviewed publications and twelve patents on the subjects of Electronic Design Automation, Reconfigurable Computing, Thermal-Aware High Performance Computing, Computer Architecture, and Instrumentation for Real-Time ML for Experimental Sciences. She is the author of the book: Heat Management in Integrated Circuits: On-chip and system-level monitoring and cooling (Materials, Circuits and Devices). Seda Ogrenci serves on the editorial boards of the IEEE Transactions on Computer Aided Design and ACM Transactions of Reconfigurable Technology and Systems.

PhD Defense in Informatics Engineering (ProDEI): ”Modular and Multi-Stage Semantic Perception System for Robotics”

Posted on

Candidate:
Bruno Georgevich Ferreira

Date, Time and Location:
27 February 2026, at 14:00, in Sala de Atos

President of the Jury:
Pedro Nuno Ferreira da Rosa da Cruz Diniz (PhD), Full Professor at the Faculdade de Engenharia da Universidade do Porto

Members:
João Alberto Fabro (PhD), Associate Professor at the Academic Department of Informatics (DAINF) of the Federal Technological University of Paraná, Brazil;
Rui Paulo Pinto da Rocha (PhD), Associate Professor at the Department of Electrical and Computer Engineering of the Faculdade de Ciências e Tecnologia da Universidade de Coimbra;
André Monteiro de Oliveira Restivo (PhD), Associate Professor at the Department of Informatics Engineering of the Faculdade de Engenharia da Universidade do Porto;
Armando Jorge Miranda de Sousa (PhD), Associate Professor at the Department of Electrical and Computer Engineering of the Faculdade de Engenharia da Universidade do Porto (Supervisor).

The thesis was co-supervised by Luís Paulo Gonçalves dos Reis (PhD), Associate Professor in the Department of Informatics Engineering of the Faculdade de Engenharia da Universidade do Porto.

Abstract:

The evolution of autonomous robotics benefits largely from the capacity to construct rich, navigable, and semantic representations of the environment, even more so if shared with humans. While the advent of open-vocabulary scene graphs powered by Vision-Language Models (VLMs) has revolutionized perception, these systems face critical hurdles: high rates of hallucinations (False Positives), a lack of topological spatial context, and operational fragility due to heavy reliance on cloud connectivity. This thesis proposes the Hybrid Inference Perception and Mapping System
(HIPaMS), framework adaptable to a target system, likely a robotic system that interacts with humans. The HIPaMS is a modular framework designed to bridge the gap between low-level perception and high-level agentic reasoning. A Proof of Concept (PoC) was designed to implement the HIPaMS. This PoC enhances the state-of-the-art ConceptGraphs semantic mapping process and introduces a refined interaction system through four main contributions. First, it introduces the Hybrid Adaptable Resource-Aware Inference Mechanism (HARAIM), which dynamically orchestrates internal models and settings based on runtime resource availability and optimization policies. This mechanism allows any optimization policy to adapt robotic system’s operation, possibly allowing zero downtime during network failures, graceful degradation and/or operational efficiency. Second, the semantic mapping pipeline is enhanced with rigorous False Positive filtering protocols, persona-based prompt engineering, and a broad collection of semantic information in an optimized manner during mapping. Third, a Room Semantic Segmentation Routine is proposed to provide topological information to the semantic map during interaction. This transforms unstructured, noisy detections into a hierarchically organized scene graph, anchoring objects within functional topological regions. Fourth, the robotic system now incorporates dynamic knowledge base via the Human-in-the-Loop (HITL) Agentic Retrieval-Augmented Generation (RAG)-based Interaction System (HARBIS). This interface uses short- and long-term memory to understand complex natural language queries. It enables the robot to learn continuously from user interactions, address gaps in perception and knowledge, maintain temporal consistency, and acknowledge its limitations by proactively asking for clarification. Extensive validation was conducted across 30 diverse environments, involving a total of 3300 interactive requests (depend on semantic map quality). The tested PoC processed 110 user requests per environment, categorized into: direct (30), indirect (30), graceful failure (30), follow-up (10) and time consistency (10). An ablation study was also performed to identify the impact of specific framework and PoC components. The results show that the PoC reduces False Positive detections by ≈86%, elevating mapping precision from a baseline of ≈ 0.28 to ≈ 0.68. Although strict filtering reduces raw recall, the integration of HITL learning increased the success rate for complex query resolution to ≈ 0.81, compared to baseline values of ≈ 0.48 and ≈ 0.55. Furthermore, the HIPaMS PoC reduced cloud inference costs by up to ≈ 84% in mapping and over ≈ 95% in interaction tasks while ensuring system stability. The presented framework pave the way for increased robotic autonomy and efficiency. The presented PoC demonstrates superior performance, particularly for human-centered scenarios.

Keywords: Semantic Mapping; Open-Vocabulary Perception; Hybrid Inference Architecture; Adaptable Framework; Human-in-the-Loop; Retrieval-Augmented Generation (RAG); Topological Segmentation; Robot@VirtualHome; Vision-Language Models; Agentic AI; Operational Robustness.

Creativity Talks | “Future Foods – Foods for the Future” by Prof. António Vicente (U.Minho)

Posted on

The 18th Creativity Talk will be given by António Vicente, Professor at the University of Minho and a nationally and internationally renowned researcher for his immense scientific contribution in the fields of Biotechnology and Food Bioengineering. In the lecture “Future Foods – Foods for the Future”, we will discover how trends such as clean label, functional foods and alternative proteins, combined with technologies such as artificial intelligence and cellular agriculture, are transforming the way we eat and produce food.
An excellent opportunity to understand how we can improve the health of people and the planet, and be part of this change!

Online broadcast on 5 March at 17:30 on the Creativity Talks YouTube channel.

The moderator will be Prof. Cláudia Gomes Silva, from the Department of Chemical and Biological Engineering of the Faculdade de Engenharia da Universidade do Porto.

About the Talk:

“There has been a significant shift in the consumers’ preferences, acceptance and needs in the last ten years, which has been particularly strong in the last five years. The “top trends” are: Clean claims (e.g. preservatives free); Clean labels; Lifestyle enhancers (e.g. high energetic foods); Functional foods (e.g. with nutraceutical function); Minimally processed foods (e.g. using natural ingredients as much as possible) and the so-called “Green foods” (making use of the benefits of plants – e.g. replacement of animal protein by other protein sources).
Along with this shift, there are two major problems related with the food we eat: I) ensuring people’s food, health and wellbeing, and II) ensuring the health of our planet.
When answering to problem I), the future food needs to tackle malnutrition, reduce calorie density, reduce food digestibility, increase micronutrient bioavailability, control gut health, allow personalized nutrition and provide appropriate food for the elderly.
In order to answer to problem II), we need to make use a set of tools for the future: molecular biology, nanotechnology, artificial intelligence, robots & sensors, the so-called “Cellular agriculture” and search for alternative protein sources.
In this talk the latest developments made by our research group towards tackling some of these challenges are going to be presented, together with our vision on what still needs to be done and which partnerships are important to lead us to the future of foods, producing foods for the future.”

About the Speaker:

António Vicente graduated in Food Engineering from the Portuguese Catholic University in 1994, received his PhD in 1998 and did his Habilitation in 2010 in Chemical and Biological Engineering from the University of Minho. He is Professor in the Department of Biological Engineering, which he directed prior to his subsequent appointments as vice-Dean of the School of Engineering and Director of the Doctoral College of that University. Currently he serves as Dean of the School of Engineering.
António Vicente is a Senior Member and Specialist in Food Engineering by the Portuguese Engineers Association.
As a researcher, he has dedicated his work to the development of micro and nanotechnological systems for application in the Agrofood sector, to the evaluation of their behavior in dynamic in vitro digestion systems, and to the study of the influence of the application of electric fields in cells and biomolecules.
He has published >380 articles in international ISI WOS journals, >30 chapters in books of international circulation, >400 papers in congresses, 5 patents and edited 5 scientific books, yielding an h-index of 95. He won the Food and Nutrition Awards in 2015 and 2017 in the R&D category. During six years (2018, 2019, 2020, 2021, 2022 and 2023) he was distinguished as Highly Cited Researcher by Clarivate Analytics and in the last five years (2020, 2021, 2022, 2023 and 2024) he has integrated the list of the top 2 % most cited researchers according to the list published by Stanford University. In 2021 he was awarded the Scientific Merit Award from the University of Minho and the Diploma of Scientific Merit from the School of Engineering of the University of Minho yearly since 2021.

DEI Talks | “The Geometry of Logic” by Prof. Cristina Videira Lopes (University of California)

Posted on

The talk entitled “The Geometry of Logic” will be presented by Prof. Cristina (Crista) Videira Lopes, from the University of California, next Wednesday, February the 25th, at 14:00, in room I-105. The session will be moderated by Prof. Rui Maranhão (DEI).

About the talk:

“Large Language Models (LLMs) exhibit surprising reasoning capabilities, yet the internal mechanisms driving these behaviors remain opaque. We hypothesize that this emergence may be driven by soft stratification: the spontaneous (and inefficient) discovery of orthogonal subspaces that separate control flow from data flow. To explore this, we introduce the STRAT architecture (STratified Registers And Types), which imposes hard stratification by explicitly partitioning the embedding space.
We evaluate STRAT on algorithmic tasks requiring precise logical manipulation. Despite having no pre-programmed knowledge of those tasks, the model spontaneously discovers interpretable
geometric topologies (e.g., antipodal operator separation) to solve the tasks. These geometric constraints also yield extreme data efficiency: models converge to the correct logical rules of arithmetic from as few as N=10 training examples. These results suggest that the stratification of the embedding space is a promising geometric substrate for neural logic.”

About the Speaker:

Cristina (Crista) Videira Lopes is a Professor of Informatics in the School of Information and Computer Sciences at the University of California, Irvine. Her research focuses on programming and software engineering for large-scale data and systems. Early in her career, she was a founding member of the team at Xerox PARC that developed Aspect-Oriented Programming. Along with her research program, she is also a prolific software developer. Her open source contributions include being one of the core developers of OpenSimulator, a virtual world server. She is also a founder and consultant of Encitra, a company specializing in online virtual reality for early-stage sustainable urban redevelopment projects. Her book “Exercises in Programming Style” has gained rave reviews, including being chosen as “Notable Book” by the ACM Best of Computing reviews. She has a PhD from Northeastern University, and MS and BS degrees from Instituto Superior Tecnico in Portugal. She is the recipient of several National Science Foundation grants, including a prestigious CAREER Award. She claims to be the only person in the world who is both an ACM Distinguished Scientist and Ohloh Kudos Rank 9.

The 23rd edition of U.Porto’s Mostra is on

Posted on

From today until Sunday, the Multiusos de Gondomar will host another edition of the U.Porto’s Mostra. Over the course of four days, visitors will be welcomed by students, lecturers and technical staff from the various faculties of the University of Porto, research centres, central services, museums and UPTEC.

Through dozens of stands, the most curious visitors will be able to try out numerous interactive activities and learn about the range of bachelor’s and master’s degree courses on offer.
Students thinking of applying for higher education, and their families, will have the opportunity to attend several information sessions, through conferences covering a variety of topics ranging from the transition from secondary to higher education to the social support available to them, as well as several information sessions.

The Informatics Engineering stand will be attended by several students from NIAEFEUP (Núcleo de Informática da Associação de Estudantes da Faculdade de Engenharia da Universidade do Porto), who will be ready to answer visitors’ questions about the Bachelor’s and Master’s degrees in Informatics and Computing Engineering, and encouraging them to try out interactive games that allow interaction with the computer through Computer Vision.

All information about the Mostra and its programme can be found on the official website. Visitors can also watch the event live online.

“EUGLOH [you·glow]: Faces behind the Alliance” features António Coelho

Posted on

The EUGLOH network has launched the series “EUGLOH [you·glow]: Faces behind the Alliance,” which features inspiring people from the EUGLOH community.

In their most recent interview, they featured António Coelho (DEI), who has been actively involved in EUGLOH since its early stages. António’s work focuses on pedagogical innovation, transversal skills, and supporting initiatives that enhance employability and lifelong learning. Through his engagement with EUGLOH, António has contributed to the design and delivery of blended intensive courses, research conferences, and other transformative initiatives that bring together students, teachers, and staff across Europe.

How has being part of EUGLOH influenced your personal and/or professional growth?

Being part of EUGLOH has been a tremendous challenge and opportunity. The Alliance has encouraged me to focus on pedagogical innovation, expanding my understanding of career services, lifelong learning, and the structures that support student employability. It has also provided an international perspective, allowing me to learn how different universities approach teaching, learning, and professional development—and to integrate the best practices from across the alliance. Personally and professionally, it has been inspiring to participate in a project that aims to shape the future of higher education in Europe.

What has been your favourite experience or moment within EUGLOH so far?

My favourite experiences have been the Blended Intensive Courses. These courses bring students, staff, and professors together to solve challenges collaboratively in just one week. It is incredible to witness the creativity and engagement of participants, and the results are often surprising and highly impactful. Another highlight has been co-creating courses, such as “Putting the students first: Designing the Classroom of the Future,” where diverse participants—professors, staff, and students—work together to design innovative learning experiences that prioritise student well-being and inclusivity. These experiences provide ideas and insights to the community that I can apply directly in my own courses, making them professionally rewarding.

What is your favourite EUGLOH initiative?

I value all initiatives that bring together students, professors, and staff to foster collaboration and learning. This includes blended courses, student research conferences, and the Annual Summit. For example, in 2024, the Annual Summit featured two Nobel Prize winners as keynote speakers, moderated by students themselves—an impressive demonstration of student engagement and excellence. These initiatives not only enhance knowledge but also promote European values, cultural exchange, and professional competencies.

From your perspective, why is EUGLOH important?

EUGLOH equips students and professionals with both technical and transversal skills, including multi-disciplinarity and multicultural understanding. By working together across countries, participants gain diverse perspectives, fostering their growth as European citizens capable of tackling global challenges. Individual universities alone may lack the scale or diversity to compete internationally, but through EUGLOH, we leverage collective strengths and collaboration. This diversity and unity make Europe a stronger, more competitive space in higher education and research, while preparing participants to be agents of positive change.

PhD Defense in Informatics Engineering (ProDEI): ”Low-Resource Machine Translation for Emakhuwa: Transfer Learning, Data Augmentation, and Lexical Resource Integration”

Posted on

Candidate:
Felermino Dário Mário António Ali

Date, Time and Location:
20 February 2026, 14:00, Sala de Atos da Faculdade de Engenharia da Universidade do Porto

President of the Jury:
Pedro Nuno Ferreira da Rosa da Cruz Diniz (PhD), Full Professor, Faculdade de Engenharia da Universidade do Porto

Members:
Maarit Tuulikki Koponen (PhD), Professor at the School of Humanities of the Philosophical Faculty of the University of Eastern Finland (Finland);
Maria Luísa Torres Ribeiro Marques da Silva Coheur (PhD), Associate Professor, Departamento de Engenharia Informática, Instituto Superior Técnico da Universidade de Lisboa;
Sérgio Sobral Nunes (PhD), Associate Professor, Departamento de Engenharia Informática, Faculdade de Engenharia da Universidade do Porto;
Henrique Daniel de Avelar Lopes Cardoso (PhD), Associate Professor, Departamento de Engenharia Informática, Faculdade de Engenharia da Universidade do Porto (Supervisor).

The thesis was co-supervised by Rui Manuel Sousa Silva (PhD), Assistant Professor at Faculdade de Letras da Universidade do Porto.

Abstract:

“This research explores the underrepresentation of low-resource languages in the field of machine translation, with a specific focus on Emakhuwa, the most widely spoken local language in Mozambique. Despite having over 7 million native speakers, Emakhuwa remains underrepresented in both academia and technology due to a lack of digital resources and linguistic tools. To fill this gap, we have developed the first significant machine translation resources for the Portuguese–Emakhuwa language pair. Our contributions include the creation of a parallel corpus through the manual translation of journalistic texts, the digitisation of existing materials, and the translation of established machine translation evaluation benchmarks. We evaluated three central strategies to improve machine translation performance in this low-resource setting: (1) transfer learning using multilingual and Africa-centred models, (2) data augmentation through back-translation, and (3) integration of external linguistic resources such as loan glossaries and bilingual dictionaries. The results show that encoder-decoder models, particularly translation-optimised architectures such as NLLB and M2M-100, perform as well as or better than larger decoder-only models while maintaining computational efficiency. Back-translation offers modest improvements, and the integration of loanwords and dictionary resources, especially in the Portuguese-Emakhuwa direction, significantly improves translation quality, especially with the use of LLMs. This work lays the foundation for future research in NLP for underrepresented languages and demonstrates practical paths for the development of machine translation systems in resource-limited contexts.”

Habilitation Exams | “Fair Resource Sharing in Concurrent Workflow Scheduling for Heterogeneous Systems” by Prof. Jorge Manuel Gomes Barbosa

Posted on

Habilitation Exams in the field of Informatics Engineering

Requested by:
Prof. Dr. Jorge Manuel Gomes Barbosa

19 February 2026, at 14:30, in FEUP’s Sala de Atos 
Assessment of the curriculum and report on the course unit “Advanced Parallel Computing”

20 February 2026, at 10:00, in FEUP’s Sala de Atos
Discussion of the synthesis seminar entitled “Fair Resource Sharing in Concurrent Workflow Scheduling for heterogeneous systems”

Chair of the Jury:
Prof. Dr. Jaime dos Santos Cardoso, Full Professor and Vice-President of the Scientific Council of Faculdade de Engenharia da Universidade do Porto

Members:
Prof. Dr. Alba Cristina Magalhaes Alves de Melo, Full Professor at the Departamento de Ciência da Computação do Instituto de Ciências Exatas da Universidade de Brasília, Brasil;
Prof. Dr. Pedro Petersen Moura Trancoso, Full Professor at the Department of Computer Science and Engineering of Chalmers University of Technology, Sweden;
Prof. Dr. Leonel Augusto Pires Seabra de Sousa, Full Professor at the Departamento de Engenharia Eletrotécnica e de Computadores do Instituto Superior Técnico da Universidade de Lisboa;
Prof. Dr. João Manuel Paiva Cardoso, Full Professor at the Departamento de Engenharia Informática da Faculdade de Engenharia da Universidade do Porto;
Prof. Dr. Carlos Miguel Ferraz Baquero-Moreno, Full Professor at the Departamento de Engenharia Informática da Faculdade de Engenharia da Universidade do Porto.

DEI Talks | “The impact of link recommendation algorithms on human social dynamics” by Prof. Fernando Santos (University of Amsterdam)

Posted on

The talk entitled “The impact of link recommendation algorithms on human social dynamics” will be presented by Prof. Fernando Pascoal dos Santos (University of Amsterdam) on March the 20th, at 11:00, in room I-105.

About the Talk:

“Online social networks increasingly shape human beliefs and behavior. In these environments, algorithms to personalize contents and provide recommendations are pervasive. Link recommendation algorithms are implemented to recommend new connections to online platforms users, based on supposed familiarity, similar interests, or the potential to serve as a source of useful information. These algorithms influence the evolution of social networks, yet their long-term impacts on human social dynamics remain unclear. In this talk, I will discuss models to study such effects. I will discuss how algorithmic link recommendations interplay with opinion dynamics, and the potential long-term impacts of such algorithms on polarization. I will also discuss methods based on agentic multi-systems, powered by LLMs, to test social media interventions aiming at mitigating polarized dynamics. We will observe that preferentially establishing links with structurally similar nodes (i.e., sharing many neighbors) results in network topologies that are amenable to opinion polarization.”

About the Speaker:

Fernando P. Santos is an Associate Professor at the University of Amsterdam. He is a member of the Socially Intelligent Artificial Systems group, where he leads the Prosocial Dynamics Lab. Fernando’s research lies at the interface of AI and Complex Systems: he is interested in understanding behavioral dynamics in systems of adaptive learning agents and designing (pro)Social AI. Previously, Fernando was a James S. McDonnell Postdoctoral Fellow at Princeton University. He completed his PhD in Computer Science and Engineering at Instituto Superior Técnico with Francisco C. Santos, Jorge M. Pacheco, and Ana Paiva. Fernando is an ELLIS Scholar and member of the board of directors of the International Foundation for Autonomous Agents and Multiagent Systems. He was awarded an ERC Starting Grant to study the impact of link-recommendation algorithms on human behavioural dynamics.

DEI Talks | “High Performance Computing for Bioinformatics Applications: the Quest for Performance” by Prof. Alba Melo

Posted on

The talk entitled “High Performance Computing for Bioinformatics Applications: the Quest for Performance” will be presented by Prof. Alba Alves de Melo (University of Brasilia) and will take place on the 20th February, at 14:30, in room B018. Prof. João Bispo (DEI) will moderate the session.

About the Talk:

“Bioinformatics applications are often computationally intensive, making High-Performance Computing (HPC) highly desirable. In this lecture, I will present parallel bioinformatics applications developed for High-Performance Computing (HPC) environments over the years by my research group at the LAICO Laboratory at the University of Brasília. The following will be addressed: (a) parallel applications for exact pairwise comparison of long biological sequences in clusters of GPUs (Graphics Processing Units) and CPUs; (b) exact multiple sequence alignment applications for multithreaded architectures; (c) exact RNA secondary structure prediction (folding and alignment) in GPU; and (d) heuristic protein folding in a supercomputer. Finally, a framework for executing scientific workflows in the HPC cloud will be presented.”

About the Speaker:

Alba Cristina Magalhaes Alves de Melo obtained her PhD in Computer Science from the Institut National Polytechnique de Grenoble (INPG), France, in 1996. Since 1997, she is with the Department of Computer Science at the University of Brasilia, Brazil, where she is now Full Professor. Prof. Melo is IEEE Senior Member, Vice-Coordinator of the IEEE Technical Community of Parallel Processing (TCPP) since 2024 and Member of the Counselling Committee in Computer Science for CNPq/Brazil since 2025. Prof. Melo received the following awards: 2023 IEEE Technical Committee on Parallel Processing (TCPP) Outstanding Service and Contributions Award; 2019 Wilkes Award for the Best Paper Published in The Computer Journal in 2018, Oxford University Press (joint work with the UPC/BSC team); 2016 Award for Advisor of the Best PhD Thesis in Computer Science in Brazil (Premio Capes de Tese).
She is Associate Editor-in-Chief of the Applications section of the Journal of Parallel and Distributed Systems (JPDC). She is also Associate Editor of many prestigious journals such as IEEE Transactions on Computers, IEEE Transactions on Parallel and Distributed Systems, ACM Computing Surveys and Future Generation Computer Systems. She was Co-General Chair of IEEE International Parallel and Distributed Processing Symposium (IPDPS) 2024. She has served as Program Chair or Track Chair of many prestigious conferences in high performance computing such as IPDPS, Supercomputing (SC), Euro-Par, HiPEAC, Cluster, ICPP, SBAC-PAD and HiPC. Prof. Melo’s research group has established long lasting collaborations with research teams from the University of Ottawa, Canada (since 2005); INRIA/Saclay and Mines Paris Tech, France (since 2011); Universitat Politecnica de Catalunya/ Barcelona Supercomputing Center (UPC/BSC), Spain (since 2012); and University of Copenhagen, Denmark (since 2017). Her research interests are high performance computing, bioinformatics and cloud computing.