Search

Human pose estimation, a vital task in computer vision, involves detecting and localising human joints in images and videos. While single-frame pose estimation has seen significant progress, it often fails to capture the temporal dynamics for understanding complex, continuous movements. We propose Poseidon, a novel multi-frame pose estimation architecture that extends the ViTPose model by integrating temporal information for enhanced accuracy and robustness to address these limitations. Poseidon introduces key innovations: (1) an Adaptive Frame Weighting (AFW) mechanism that dynamically prioritises frames based on their relevance, ensuring that the model focuses on the most informative data; (2) a Multi-Scale Feature Fusion (MSFF) module that aggregates features from different backbone layers to capture both fine-grained details and high-level semantics; and (3) a Cross-Attention module for effective information exchange between central and contextual frames, enhancing the model's temporal coherence. The proposed architecture improves performance in complex video scenarios and offers scalability and computational efficiency suitable for real-world applications. Our approach achieves state-of-the-art performance on the PoseTrack21 and PoseTrack18 datasets, achieving mAP scores of 88.3 and 87.8, respectively, outperforming existing methods.

Maintaining software quality is crucial in the dynamic landscape of software development. Regression testing ensures that software works as expected after changes are implemented. However, re-executing all test cases for every modification is often impractical and costly, particularly for large systems. Although very effective, traditional test suite optimization techniques are often impractical in resource-constrained scenarios, as they are computationally expensive. Hence, quantum computing solutions have been developed to improve their efficiency but have shown drawbacks in terms of effectiveness. We propose reformulating the regression test case selection problem to use quantum computation techniques better. Our objectives are (i) to provide more efficient solutions than traditional methods and (ii) to improve the effectiveness of previously proposed quantum-based solutions. We propose SelectQA, a quantum annealing approach that can outperform the quantum-based approach BootQA in terms of effectiveness while obtaining results comparable to those of the classic Additional Greedy and DIV-GA approaches. Regarding efficiency, SelectQA outperforms DIV-GA and has similar results with the Additional Greedy algorithm but is exceeded by BootQA.

Horizon 2020 and Horizon Europe the EU programs supporting research and innovation through collaboration between companies, academic institutions, and research organisations. This paper introduces a novel methodology using open data on Horizon programs to analyse collaborations, leadership roles, and their evolution, with a focus on the North Adriatic Hydrogen Valley project in the hydrogen energy sector. The methodology employs network analysis, transforming tabular data into weighted networks that represent collaborations between organisations. Centrality measures and community detection algorithms identify influential organisations and stable partnerships over time. To ensure robust and reliable results, the methodology addresses challenges such as input-ordering bias and result variability, while the exploration of the solution space enhances the accuracy of identified collaboration patterns. The case study reveals key leaders and stable communities within the hydrogen energy sector, providing valuable insights for policymakers and organisations fostering innovation through sustained collaborations. The proposed methodology effectively identifies influential organisations and tracks the stability of research collaborations. The insights gained are valuable for policymakers and organisations seeking to foster innovation through sustained partnerships. This approach can be extended to other sectors, offering a framework for understanding the impact of EU research funding on collaboration and leadership dynamics.

This article explores the importance of examining the solution space in community detection, highlighting its role in achieving reliable results when dealing with real-world problems. A Bayesian framework is used to estimate the stability of the solution space and classify it into categories Single, Dominant, Multiple, Sparse or Empty. By applying this approach to real-world networks, the study highlights the importance of considering multiple solutions rather than relying on a single partition. This ensures more reliable results and efficient use of computational resources in community detection analysis.

Clickbait PDFs, an entry point for multiple Web attacks, are distributed via SEO poisoning and rank high in search results due to being massively uploaded on abused or compromised websites. The central role of these hosts in the distribution of clickbait PDFs remains understudied, and it is unclear whether attackers differentiate the types of hosting for PDF uploads, how long they rely on hosts, and how affected parties respond to abuse. To address this, we conducted real-time analyses on hosts, collecting data on 4,648,939 clickbait PDFs served by 177,835 hosts over 17 months. Our results revealed a diverse infrastructure, with hosts falling into three main hosting types. We also identified at scale the presence of eight software components which facilitate file uploads and which are likely exploited for clickbait PDF distribution. We contact affected parties to report the misuse of their resources via a large-scale vulnerability notification. While we observed some effectiveness in terms of number of cleaned-up PDFs following the notification, long-term improvement in this infrastructure remained insignificant. This finding raises questions about the hosting providers' role in combating abuse and the actual impact of vulnerability notifications., Comment: Published at Euro S&P 2024

Retrieval Augmented Generation (RAG) is a technique commonly used to equip models with out of distribution knowledge. This process involves collecting, indexing, retrieving, and providing information to an LLM for generating responses. Despite its growing popularity due to its flexibility and low cost, the security implications of RAG have not been extensively studied. The data for such systems are often collected from public sources, providing an attacker a gateway for indirect prompt injections to manipulate the responses of the model. In this paper, we investigate the security of RAG systems against end-to-end indirect prompt manipulations. First, we review existing RAG framework pipelines, deriving a prototypical architecture and identifying critical parameters. We then examine prior works searching for techniques that attackers can use to perform indirect prompt manipulations. Finally, we implemented Rag 'n Roll, a framework to determine the effectiveness of attacks against end-to-end RAG applications. Our results show that existing attacks are mostly optimized to boost the ranking of malicious documents during the retrieval phase. However, a higher rank does not immediately translate into a reliable attack. Most attacks, against various configurations, settle around a 40% success rate, which could rise to 60% when considering ambiguous answers as successful attacks (those that include the expected benign one as well). Additionally, when using unoptimized documents, attackers deploying two of them (or more) for a target query can achieve similar results as those using optimized ones. Finally, exploration of the configuration space of a RAG showed limited impact in thwarting the attacks, where the most successful combination severely undermines functionality.

Complex data in social and natural sciences find effective representation through networks, wherein quantitative and categorical information can be associated with nodes and connecting edges. The internal structure of networks can be explored using unsupervised machine learning methods known as community detection algorithms. The process of community detection is inherently subject to uncertainty as algorithms utilize heuristic approaches and randomised procedures to explore vast solution spaces, resulting in non-deterministic outcomes and variability in detected communities across multiple runs. Moreover, many algorithms are not designed to identify outliers and may fail to take into account that a network is an unordered mathematical entity. The main aim of our work is to address these issues through a consensus-based approach by introducing a new framework called Consensus Community Detection (CCD). Our method can be applied to different community detection algorithms, allowing the quantification of uncertainty for the whole network as well as for each node, and providing three strategies for dealing with outliers: incorporate, highlight, or group. The effectiveness of our approach is evaluated on artificial benchmark networks., Comment: 22 pages, 11 figures. Submitted to Machine Learning

This study presents an innovative method for Alzheimer's disease diagnosis using 3D MRI designed to enhance the explainability of model decisions. Our approach adopts a soft attention mechanism, enabling 2D CNNs to extract volumetric representations. At the same time, the importance of each slice in decision-making is learned, allowing the generation of a voxel-level attention map to produce an explainable MRI. To test our method and ensure the reproducibility of our results, we chose a standardized collection of MRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). On this dataset, our method significantly outperforms state-of-the-art methods in (i) distinguishing AD from cognitive normal (CN) with an accuracy of 0.856 and Matthew's correlation coefficient (MCC) of 0.712, representing improvements of 2.4% and 5.3% respectively over the second-best, and (ii) in the prognostic task of discerning stable from progressive mild cognitive impairment (MCI) with an accuracy of 0.725 and MCC of 0.443, showing improvements of 10.2% and 20.5% respectively over the second-best. We achieved this prognostic result by adopting a double transfer learning strategy, which enhanced sensitivity to morphological changes and facilitated early-stage AD detection. With voxel-level precision, our method identified which specific areas are being paid attention to, identifying these predominant brain regions: the hippocampus, the amygdala, the parahippocampal, and the inferior lateral ventricles. All these areas are clinically associated with AD development. Furthermore, our approach consistently found the same AD-related areas across different cross-validation folds, proving its robustness and precision in highlighting areas that align closely with known pathological markers of the disease., Comment: 21 pages, 9 figures, 9 tables

Alzheimer's disease is one of the most incisive illnesses among the neurodegenerative ones, and it causes a progressive decline in cognitive abilities that, in the worst cases, becomes severe enough to interfere with daily life. Currently, there is no cure, so an early diagnosis is strongly needed to try and slow its progression through medical treatments. Handwriting analysis is considered a potential tool for detecting and understanding certain neurological conditions, including Alzheimer's disease. While handwriting analysis alone cannot provide a definitive diagnosis of Alzheimer's, it may offer some insights and be used for a comprehensive assessment. The Sigma-lognormal model is conceived for movement analysis and can also be applied to handwriting. This model returns a set of lognormal parameters as output, which forms the basis for the computation of novel and significant features. This paper presents a machine learning approach applied to handwriting features extracted through the sigma-lognormal model. The aim is to develop a support system to help doctors in the diagnosis and study of Alzheimer, evaluate the effectiveness of the extracted features and finally study the relation among them.

Using kinematic properties of handwriting to support the diagnosis of neurodegenerative disease is a real challenge: non-invasive detection techniques combined with machine learning approaches promise big steps forward in this research field. In literature, the tasks proposed focused on different cognitive skills to elicitate handwriting movements. In particular, the meaning and phonology of words to copy can compromise writing fluency. In this paper, we investigated how word semantics and phonology affect the handwriting of people affected by Alzheimer's disease. To this aim, we used the data from six handwriting tasks, each requiring copying a word belonging to one of the following categories: regular (have a predictable phoneme-grapheme correspondence, e.g., cat), non-regular (have atypical phoneme-grapheme correspondence, e.g., laugh), and non-word (non-meaningful pronounceable letter strings that conform to phoneme-grapheme conversion rules). We analyzed the data using a machine learning approach by implementing four well-known and widely-used classifiers and feature selection. The experimental results showed that the feature selection allowed us to derive a different set of highly distinctive features for each word type. Furthermore, non-regular words needed, on average, more features but achieved excellent classification performance: the best result was obtained on a non-regular, reaching an accuracy close to 90%.

Context. Quantum computing is becoming a reality, and quantum software engineering (QSE) is emerging as a new discipline to enable developers to design and develop quantum programs. Objective. This paper presents a systematic mapping study of the current state of QSE research, aiming to identify the most investigated topics, the types and number of studies, the main reported results, and the most studied quantum computing tools/frameworks. Additionally, the study aims to explore the research community's interest in QSE, how it has evolved, and any prior contributions to the discipline before its formal introduction through the Talavera Manifesto. Method. We searched for relevant articles in several databases and applied inclusion and exclusion criteria to select the most relevant studies. After evaluating the quality of the selected resources, we extracted relevant data from the primary studies and analyzed them. Results. We found that QSE research has primarily focused on software testing, with little attention given to other topics, such as software engineering management. The most commonly studied technology for techniques and tools is Qiskit, although, in most studies, either multiple or none specific technologies were employed. The researchers most interested in QSE are interconnected through direct collaborations, and several strong collaboration clusters have been identified. Most articles in QSE have been published in non-thematic venues, with a preference for conferences. Conclusions. The study's implications are providing a centralized source of information for researchers and practitioners in the field, facilitating knowledge transfer, and contributing to the advancement and growth of QSE.