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.

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.”

Candidate:
Ivone Manuela Neiva Santos

Date, Time and Location:
29 January 2026, 14:30, Sala de Atos da Faculdade de Engenharia da Universidade do Porto

President of the Jury:
António Fernando Vasconcelos Cunha Castro Coelho (PhD), Associate Professor with Habilitation from Faculdade de Engenharia, Universidade do Porto

Members:
Marisa Rodrigues Pinto Torres da Silva (PhD), Full Professor, Faculdade de Ciências Sociais e Humanas, Universidade Nova de Lisboa;
Maria Madalena da Costa Oliveira (PhD), Associate Professor, Instituto de Ciências Sociais, Universidade do Minho;
Maria José Lisboa Brites de Azeredo (PhD), Associate Professor with Habilitation, Faculdade de Comunicação, Arquitetura, Artes e Tecnologias da Informação, Universidade Lusófona;
Ana Isabel Crispim Mendes Reis (PhD), Associate Professor, Department of Communication and Information Sciences, Faculdade de Letras, Universidade do Porto (Supervisor);
Ricardo José Pinheiro Fernandes Morais (PhD), Assistant Professor, Department of Communication and Information Sciences, Faculdade de Letras, Universidade do Porto.

The thesis was co-supervised by José Manuel Pereira Azevedo (PhD), Full Professor, Department of Communication and Information Sciences, Faculdade de Letras, Universidade do Porto.

Abstract:

Empathy, which is defined as the ability to understand and share the emotional state of others, is considered to be fundamental to both personal well-being and social cohesion. Its presence has been linked to pro-social behaviour, while its absence has been associated with a greater predisposition to aggressive behaviour. Empathy is a multidimensional construct integrating affective and cognitive components, and educational interventions appear to positively influence its development. While cognitive empathy is generally considered to be more sensitive to education, the affective component appears to benefit from emotional activities. Empathy is now considered a vital skill for the ‘digital citizen’, but research suggests that empathy displayed online is lower than empathy displayed offline. Tendencies such as inattention, desensitisation and disinhibition, which are stimulated by the internet, seem to make empathy more difficult to achieve. The fact that empathic capacity primarily develops towards the end of adolescence highlights the importance of exploring strategies to foster it throughout education, particularly in an era where digital technology is becoming increasingly prevalent in all areas of social life. This scenario highlights the need to deepen our understanding of digital empathy and consider strategies for promoting it in education. This is reflected in the objectives that guided this thesis. The research underlying this thesis includes a review of the literature on empathy and the methodologies employed to study and encourage it. It features a critical analysis of the role of screens in young people’s daily lives, as well as of the different approaches to the relationship between empathy and digital technology. Given that sound is a privileged vehicle for emotional connection with characteristics that make it resilient to digital environments, the review also explores the literature on the potential of auditory stimuli and audio narratives to promote empathy. Supported by this review, the empirical research comprises two studies: a descriptive study and a quasi-experimental study. These studies involved three educational institutions at different levels and students in the adolescent age group (10–24 years). A total of 279 students participated in the descriptive study and 228 in the quasi-experimental study, of whom 76 were in the experimental group. The descriptive study measured and compared participants’ empathy and digital empathy using a questionnaire based on self-report scales previously used in empathy research with this age group. The quasi-experimental study assessed the impact of an educational programme designed to explore the potential of sound and narrative. Based on the experience-based learning model, the programme combined technical and socio-emotional learning through Media Education. It is organised into two modules. The first module consists of a set of group dynamics exploring the theme of empathy and its relationship with digital environments and auditory stimuli. The second module considers the process of producing audio narratives with emotional content. The intervention’s impact was assessed both quantitatively, via pre- and post-test surveys, and qualitatively, through analysis of the narratives and other texts produced by participants throughout the programme. Overall, the results of the descriptive study indicate that digital empathy is lower than general empathy, with the affective component being lower than the cognitive component on both scales. The results also show that girls have higher levels of both empathy and digital empathy. Age appears to be a differentiating factor in empathy levels, but not in digital empathy. Results suggest that digital empathy does not increase significantly during adolescence, unlike general empathy. These results therefore support the need for educational interventions to stimulate empathy from early adolescence onwards addressing its multidimensionality and various contexts. The quantitative impact of participation in the educational programme evaluated in the quasi-experimental study was not significant. Nevertheless, a qualitative analysis of the data suggests that participation in the programme provided an opportunity to experiment with different empathic practices. The programme can therefore be considered a tool that facilitates the reconciliation of technical learning with the development of empathy. This tool can be applied to different stages of adolescence, levels of education, and school contexts. The findings of this research reiterate the concerns expressed in existing literature about the impact of digital environments on empathy development among young people. The 0findings suggest that programmes based on producing audio narratives with emotional content could promote empathy in educational contexts, addressing the constraints imposed by digital environments. Based on these findings, this research has produced a manual to disseminate the tested educational model and a validated instrument to measure empathy and digital empathy in Portuguese. To our knowledge, this is the first questionnaire of its kind to enhance sound stimuli.

Keywords: digital empathy; audio production; narrative; education.

Candidate:
Acilon Himercírio Baptista Cavalcante

Date, Time and Location:
26 January 2026, 14:30, Room Professor Joaquim Sarmento (G129), Department of Civil and Georesources Engineering, Faculdade de Engenharia da Universidade do Porto

President of the Jury:
António Fernando Vasconcelos Cunha Castro Coelho (PhD), Associate Professor with Habilitation, Faculdade de Engenharia da Universidade do Porto

Members:
Isabel Alexandra Reis Gonçalves Ferreira (PhD), Researcher at the Centre for Social Studies, Universidade de Coimbra;
Ivone Marília Carinhas Ferreira (PhD), Assistant Professor, Department of Communication Sciences, Faculdade de Ciências Sociais e Humanas, Universidade Nova de Lisboa;
Ana Isabel Barreto Furtado Franco de Albuquerque Veloso (PhD), Full Professor, Department of Communication and Art, Universidade de Aveiro;
José Manuel Pereira Azevedo (PhD), Full Professor, Department of Communication and Information Sciences, Faculdade de Letras, Universidade do Porto (Supervisor);
Maria Van Zeller de Macedo de Oliveira e Sousa (PhD), Invited Assitant Professor, Department of Informatics Engineering, Faculdade de Engenharia, Universidade do Porto and Researcher at the Instituto de Engenharia de Sistemas e Computadores, Tecnologia e Ciência (INESC TEC).

Abstract:

This thesis investigates the promotion of social well-being in cities through the concept of urban-digital hybridity, which considers social and spatial interactions—whether physical and/or digital—as inseparable in the urban context. Based on an integrative literature review, a set of indicators was identified and categorised to more comprehensively assess the effectiveness of public policies aimed at improving quality of life in technology-mediated urban environments.
The review of indicators combined traditional methodologies—such as those used in the World Happiness Report, published by the United Nations—with metrics related to physical and mental health, community participation, perception of safety, and cultural vitality, while also incorporating emerging variables derived from the use of digital media. The research methodology adapted the mapping and critical analysis of these indicators to Marichela Sepe’s Cartography of Happiness, applying it to contexts of urban-digital hybridity and combining it with empirical digital placemaking experiments.
Case studies and digital placemaking experiences were conducted in the cities of Porto and Póvoa de Varzim, involving local communities, religious institutions, and schools, exploring technological mediation as a catalyst for social bonds and the activation of public spaces. Heatmaps of interactions, together with qualitative field data, allowed the identification of correlations between patterns of urban activation, city morphology, and landscape.
As its main outcome, the research proposes three core metrics for assessing social well-being in hybrid cities: Sense of Belonging, Sense of Place, and Sense of Community, analysed in their urban, digital, and hybrid dimensions.
The thesis’ main contribution is an integrated model for assessing urban social well-being, combining physical and digital metrics to provide an operational framework for urban planning and public policy design, aiming to foster more inclusive, participatory, and well-being-oriented cities.

Candidate:
Tiago Filipe Mendes Neves

Date, Time and Location:
16 September 2025, 15h30, Sala de Atos, Faculdade de Engenharia da Universidade do Porto

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

Members:
Keisuke Fujii (PhD), Associate Professor, Department of Intelligent Systems, Graduate School of Informatics of the Nagoya University, Japan;
Jesse Jon Davis (PhD), Full Professor, Department of Computer Science, Faculty of Engineering Science, Katholieke Universiteit Leuven, Belgium;
Luís Paulo Gonçalves dos Reis (PhD), Associate Professor with Habilitation, Departament of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto;
João Pedro Carvalho Leal Mendes Moreira (PhD), Associate Professor, Departament of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto (Supervisor).

The thesis was co-supervised by Luís Jorge Machado da Cunha Meireles (PhD), Senior Psychologist & Data Scientist, FC Porto.

Abstract:

Self-supervised large models that disrupt domains such as language, vision, and biology are transforming the world. However, these generative models that learn the underlying data distribution do not perform at the same level on all tasks. For example, Large Language Models (LLMs) do not yet have concrete applicability in soccer analytics. The models lack reasoning capabilities to provide concrete and actionable insights that can compete with the wide range of case-specific metrics within soccer analytics. While there have been some studies exploring the applicability of generative models in soccer, no study aimed for the moonshot of building a complete self-supervised learning model for soccer event data. Let’s consider the individual events (each shot, pass, tackle, …) in a soccer match the “words” that describe what is happening. We can consider each possession a “sentence,” each game an “essay,” and event data as a whole a “language.” By working within this framework, we have all the tools to build a self-supervised model in the same image as LLMs. The goal of this thesis is to build a foundation self-supervised model for soccer event data – termed Large Events Model (LEM) – and demonstrate its real-world applicability and generality in solving a wide range of tasks, such as simulation and modeling, that would otherwise require multiple different approaches. We propose three approaches to building LEMs: a chain of classifiers, causal mask modeling, and sequential language modeling with transformers. First, the chain of classifiers provides the first generative model that models all aspects of event data without posing restrictions on event types, reaching a level of performance that allows large-scale simulation of soccer matches. Then, we investigate two alternative approaches to remove some of the constraints of the first approach. The causal mask modeling approach using multilayer perceptrons reaches the state-of-the-art performance of several of our proposed benchmarks, providing a set of application-ready models to solve a wide range of soccer analytics tasks. We explore a wide range of applications, from automated strategy search with reinforcement learning to risk-reward behaviors of soccer players. More than a dozen use cases for LEMs are present in this thesis. The implications of our work are far-reaching. LEMs have the potential to become the operating system for event data in soccer analytics. They will transform the way clubs work, with easier access to machine learning models that would otherwise require tremendous modeling effort. With LEMs, the barrier to entry will lower significantly as any club in the world can access a model capable of solving its most relevant problems.

Keywords: generative models; foundation models; sports analytics; deep learning applications; simulation; soccer.

Candidate:
Gabriel de Jesus

Date, Time and Location:
1 September 2025, 14:30, Sala de Atos, Faculdade de Engenharia da Universidade do Porto

President of the Jury:
Rui Filipe Lima Maranhão de Abreu (PhD), Full Professor, Departament of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto

Members:
Arjen P. de Vries (PhD), Full Professor at the Institute for Computing and Information Sciences of the Radboud Universiteit, Nimega, The Netherlands;
Bruno Emanuel da Graça Martins (PhD), Associate Professor, Departament of Electrical and Computer Engineering, Instituto Superior Técnico da Universidade de Lisboa;
Henrique Daniel de Avelar Lopes Cardoso (PhD), Associate Professor, Departament of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto;
Sérgio Sobral Nunes (PhD), Associate Professor, Departament of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto (Supervisor).

Abstract:

Ensuring access to information in all languages is crucial for bridging disparities in communities’ participation in the digital age and fostering a more inclusive and equitable society, particularly for speakers of low-resource languages. However, enabling such access remains a significant challenge for many of these communities. Tetun, a language that transitioned from a dialect to one of Timor-Leste’s official languages when the country restored its independence in 2002, faces similar challenges. According to the 2015 census, Tetun is spoken by approximately 79% of the country’s 1.18 million population. Despite its official status, Tetun remains underserved in language technology. Specifically, information retrieval-based solutions for the language do not exist, making it challenging to find relevant information on the internet and digital platforms for text-based search in Tetun.
This work tackles these challenges by investigating retrieval strategies for text-based search that can enable the application of information retrieval techniques to develop search solutions for Tetun, with a specific focus on the ad-hoc text retrieval task. Given that language-specific algorithms, tools, and document collections for Tetun were previously unavailable, this work began by creating these foundational resources, which serve as contributions relevant to information retrieval and natural language processing domains. These resources include a tokenizer, a language identification model, a stemmer, a stopword list, a document collection, a test collection, baselines for the ad-hoc text retrieval task, and a search log dataset. The contributions to information retrieval for low-resource languages include: (1) A data collection pipeline tailored for low-resource languages to streamline the construction of textual data from the web; (2) A human-in-the-loop methodology for annotating, processing, and constructing a dataset well-suited for a variety of information retrieval and natural language processing tasks; (3) A novel network-based approach for stopword detection; (4) Methodologies for developing a stemmer, designed for a language heavily influenced by loanwords, and the construction of a ground truth set for evaluating stemmer performance; (5) A detailed approach for constructing a test collection to evaluate the effectiveness of retrieval systems; (6) A methodology for establishing a robust baseline for the ad-hoc text retrieval task; and (7) Document contextualization and dual-parameter tuning strategies for hybrid text retrieval. The results from this work contribute to the development of technologies associated with the computational processing of Tetun, address gaps in its linguistic resources, and achieve impactful outcomes that elevate Tetun’s status. These advancements open new opportunities for future research and innovation. Moreover, this work introduces promising methodologies that can be adapted to other languages facing similar challenges, thereby contributing to the broader advancement of information retrieval for low-resource languages.

Candidate: Martin Joseph Aubard

Date, time and location:
25 July 2025, 14:00, Sala de Atos DEEC – I-105, Faculty of Engineering, University of Porto

President of the Jury:
Pedro Nuno Ferreira da Rosa da Cruz Diniz (PhD), Full Professor, Department of Informatics Engineering, Faculty of Engineering, University of Porto

Members:
Bilal Wehbe (PhD), Senior Researcher at the German Research Center for Artificial Intelligence, Germany;
Catarina Helena Branco Simões da Silva (PhD), Associate Professor, Department of Computer Engineering, Faculty of Science and Technology, University of Coimbra;
Andry Maykol Gomes Pinto (PhD), Associate Professor, Department of Electrical and Computer Engineering, Faculty of Engineering, University of Porto;
Ana Maria Dias Madureira Pereira (PhD), Coordinating Professor with Aggregation, Department of Computer Engineering, Instituto Superior de Engenharia do Porto, Instituto Polítécnico do Porto (Supervisor).

The thesis was co-supervised by Luís Filipe Pinto de Almeida Teixeira (PhD), Associate Professor in the Department of Informatics Engineering at the Faculty of Engineering of the University of Porto.

Abstract:

This thesis addresses the challenge of improving Autonomous Underwater Vehicles (AUVs) onboard detection and interaction capabilities using Side-Scan Sonar (SSS) data. Traditionally, underwater missions relied on pre-defined plans where data are analyzed post-mission by operators or experts. This workflow is time-consuming, often requiring multiple missions to identify and localize underwater targets. The need for repeated missions increases operational costs and complexity, highlighting the inefficiency of current methodologies. Moreover, such approaches do not allow the AUV to interact with detected targets in real time, limiting the scope of mission adaptation and real-time decision-making. To overcome these limitations, this thesis presents a novel framework integrating deep learning models for object detection directly onboard AUVs. This integration enables the vehicle to detect, localize, and interact with underwater targets in real time, offering significant improvements over traditional post-mission analysis. The framework builds upon the LSTS toolchain, which is responsible for AUV motion control and communication, and introduces enhanced real-time data processing capabilities. However, implementing such a model into an embedded system suffers from computational limitations affecting the model’s performance. Thus, the knowledge distillation methods have been implemented, ensuring smaller, more efficient models to perform onboard detection without sacrificing accuracy. Additionally, to improve the model’s robustness against underwater noise, a novel adversarial retraining framework, ROSAR, is introduced, ensuring reliable operation even in noisy sonar environments. Following the onboard detection and localization enhancement, we focused on onboard interaction with the detected object. This is realized by extending the previous onboard framework and validating it through a customized simulator, enhancing interaction with the detected objects, and validating through a pipeline inspection use case, which reduces mission time by combining sonar detection and camera data collection in a single mission, utilizing behavior trees and safety-assessed models. Given the lack of open-source sonar datasets in the field, this thesis contributes to two novel publicly available side-scan sonar datasets, SWDD and Subpipe, which include field-collected data on walls and pipelines and are manually annotated for object detection. By shifting from post-mission analysis to real-time detection and interaction, this thesis significantly improves the operational efficiency of AUV missions. The proposed framework streamlines underwater operations and enhances AUVs’ autonomous behavior, relying on efficient, accurate, and robust object detection model for efficient underwater exploration and monitoring applications.