LAW-NLP aims to explore the use of natural language processing to help users understand and identify relevant Portuguese laws that apply to their situation. To that end, the applicant will explore the use of Neuro-Symbolic approach to create a system capable of answering natural language questions about the legality of performing an action. It uses Abstract Meaning Representation for Natural Language understanding and Knowledge Graph reasoning to infer the possibility of legally performing such action.

The grant holder will work on the final stages of development and evaluation of the Hub, a safeguarded and monitored social space which seeks to improve social well-being and positive reinforcement practices between peers by design. We expect the grant holder to detect and solve potential issues or bugs with the current system, and evaluate the Hub's feasability with 11-14-year-old participants.

The student will join the project DIVINA, Detecting Injection Vulnerabilities in Node.js Applications, in the scope of its fourth task, whose main goal is to build various benchmarks for the evaluation of detection tools for injection vulnerabilities. The student will extend a dataset consisting of a large number of Node.js packages that have been flagged as containing injection vulnerabilities. In particular, the student will generate, for each vulnerable Node.js package, the information required for synthesizing the corresponding exploit through the use of a symbolic execution tool for JavaScript.

IntelligentCare project develops a multimorbidity management decision support system focused on the patient's vision of value, i.e., achieving the outcomes that matter for patients. To this end, analytical methodologies to identify high risk multimorbidity patient groups, as well as to eval-uate and define evidence-based clinical pathways associated with each patient group will be used. The project contributes to improved healthcare delivery through precision and personalized medicine and to contribute to market sustainability using new technologies. INESC-ID leads a project activity for identifying and characterizing the main clusters of multimorbidity in clinical data. The applicant to be hired will join a team engaged in performing exploratory analysis of clinical data for clinical phenotyping and identification of most frequent representative groups in multimorbidity, and investigation and exploration of different data mining algorithms supporting this task.

The candidate will work on Tasks 2-3 of the project, addressing the problem of model-based reinforcement learning in non-stationary environments using deep learned models. The key objectives are: 1. Conduct an extensive survey of state-of-the-art on model-based reinforcement learning in non-stationary environments, where non stationarity occurs from the presence of other agents in the environment, and planning using the learned models. 2. Develop new methods for learning disentangled latent representations for such processes that can be used to enable the emergence of communication. 3. Extend MCTS planning algorithms to handle the models learned in 2.

The candidate will work on Tasks 2-3 of the project, addressing the problem of model-based reinforcement learning in non-stationary environments using deep learned models. The key objectives are: 1. Conduct an extensive survey of state-of-the-art on model-based reinforcement learning in non-stationary environments, where non stationarity occurs from the presence of other agents in the environment, and planning using the learned models. 2. Develop new methods for learning disentangled latent representations for such processes that can be used to enable the emergence of communication. 3. Extend MCTS planning algorithms to handle the models learned in 2

kNOw HATE combines social and language sciences with cutting edge computer science techniques, offering a comprehensive, interdisciplinary approach to: analyse content and propagation of online hate speech (OHS) and counter speech in user generated content (e.g., tweets) in Portugal; create models for its automatic detection; offer culturally sensitive and tailored guidelines and tools for effective detection, prevention, and countering narratives to be used by individuals, civil society organizations, and policy makers. The work to be carried out by the selected candidates has the following objectives: Linguistic analysis and annotation of comments collected from different social media platforms, such as Twitter. The corpus annotation will build upon detailed guidelines designed by the project’s team, aiming at identifying a variety of linguistic and discursive aspects associated with the materialization of hate speech and counter-speech in social media corpora.

The accepted candidates will work on improving the interoperability between C++ and Rust through static analysis techniques. There will be a strong focus on scalability, with the goal of scaling to very large codebases.

The main objective of this contract is to carry out research on efficient algorithms, architectures and tools specifically designed for maximising the performance and energy efficiency of emerging HPC systems. Special emphasis is given to sparse matrix, tensor and graph computations, while also opening up new application areas in data analytics and deep learning. This position is open in the scope of the European Project Sparcity (http://sparcity.eu/ - Euro-HPC project No 956213). Opportunities for performing high quality research, as part of a well known and highly motivated consortium of international researchers and students. Possibilities to collaborate with international industry and academic partners, and perform research visits to other universities or labs. For more information go to: https://www.euraxess.pt/jobs/797038

The main objective of this contract is to carry out research on efficient algorithms, architectures and tools specifically designed for maximising the performance and energy efficiency of emerging HPC systems. Special emphasis is given to sparse matrix, tensor and graph computations, while also opening up new application areas in data analytics and deep learning. This position is open in the scope of the European Project Sparcity (http://sparcity.eu/ - Euro-HPC project No 956213). Opportunities for performing high quality research, as part of a well-known and highly motivated consortium of international researchers and students. Possibilities to collaborate with international industry and academic partners, and perform research visits to other universities or labs. For more information go to: https://www.euraxess.pt/jobs/797036

The selected candidate will be a member of the research project GOLEM, which aims to automate programming and revolutionize the software development experience, ultimately allowing citizens or non-expert business professionals to develop their own apps. The project will advance the state-of-the-art AI applied to code, program synthesis, and automated programming techniques. It is often the case that programmers know how to structure the program that they intend to write, but they do not immediately know how to write certain sub-expressions (i.e., how to write sub-programs that are part of the main program). With this in mind, we are interested in exploring the use of program synthesis to aid the programmer in filing these unknown sub-programs. The expectation is that this will increase productivity and reduce the presence of bugs. We will focus on a functional programming setting, so that we can leverage recent developments on "typed holes". The selected candidate will: 1) Investigate approaches to fill in "typed holes" in functional programs with results obtained from existing program synthesizers 2) Investigate the limitations of the program synthesizers identified in 1) and suggest extensions that could improve the productivity of programmers 3) Produce a technical report with all the research findings

IntelligentCare project develops a multimorbidity management decision support system focused on the patient's vision of value, i.e., achieving the outcomes that matter for patients. To this end, analytical methodologies to identify high risk multimorbidity patient groups, as well as to evaluate and define evidence-based clinical pathways associated with each patient group will be used. The project contributes to improved healthcare delivery through precision and personalized medicine and to contribute to market sustainability using new technologies. INESC-ID leads a project activity for identifying and and characterizing the main clusters of multimorbidity in clinical data. The applicant to be hired will join a team engaged in performing exploratory analysis of clinical data for clinical phenotyping and identification of most frequent representative groups iin multimorbidity, and investigation and exploration of different data mining algorithms supporting this task.

Function-as-a-Service (FaaS) is a promising new cloud computing model that is becoming extremely popular, not only for running simple website requests but also for processing complex data analytics tasks, traditionally labelled as Big Data applications. FaaS applications are predominantly developed in high-level languages such as Java and, as consequence, naturally suffer from all the well-known problems related to Garbage Collection, code warmup, etc. However, FaaS imposes new restrictions on many aspects compared to traditional Big Data applications deployed in VMs. These restrictions include smaller task partition, lower compute power available to each worker, tighter time limits for each task, among others. The goal of this task is to evaluate how novaVM behaves in such FaaS environments and to point out its benefits compared to off-the-shelf runtimes. It is also a goal to point out potential areas of future improvement in novaVM so that it could adapt better to this new cloud computing model.