Search

The LHCb Ring-Imaging Cherenkov detectors are built to provide charged hadron identification over a large range of momentum. The upgraded detectors are also capable of providing an independent measurement of the luminosity for the LHCb experiment during LHC Run 3. The modelling of the opto-electronics chain, the application of the powering strategy during operations, the calibration procedures and the proof of principle of a novel technique for luminosity determination are presented. In addition, the preliminary precision achieved during the 2023 data-taking year for real-time and offline luminosity measurements is reported.

The COVID-19 pandemic has profoundly affected the normal course of life -- from lock-downs and virtual meetings to the unprecedentedly swift creation of vaccines. To halt the COVID-19 pandemic, the world has started preparing for the global vaccine roll-out. In an effort to navigate the immense volume of information about COVID-19, the public has turned to social networks. Among them, $\mathbb{X}$ (formerly Twitter) has played a key role in distributing related information. Most people are not trained to interpret medical research and remain skeptical about the efficacy of new vaccines. Measuring their reactions and perceptions is gaining significance in the fight against COVID-19. To assess the public perception regarding the COVID-19 vaccine, our work applies a sentiment analysis approach, using natural language processing of $\mathbb{X}$ data. We show how to use textual analytics and textual data visualization to discover early insights (for example, by analyzing the most frequently used keywords and hashtags). Furthermore, we look at how people's sentiments vary across the countries. Our results indicate that although the overall reaction to the vaccine is positive, there are also negative sentiments associated with the tweets, especially when examined at the country level. Additionally, from the extracted tweets, we manually labeled 100 tweets as positive and 100 tweets as negative and trained various One-Class Classifiers (OCCs). The experimental results indicate that the S-SVDD classifiers outperform other OCCs.

We describe a framework and tool specification that represents a step towards cybersecurity testing and monitoring of IoT ecosystems. We begin with challenges from a previous paper and discuss an integrated approach and tools to enable testing and monitoring to address these challenges. We also describe exemplary use cases of IoT ecosystems and propose approaches to address the challenges using the framework and tools. The current status of this work is that the specification and conceptualisation is complete, use cases are understood with clear challenges and implementation / extension of the tools and framework is underway with tools at different stages of development. Several key observations have been made throughout this work, as follows. 1) Tools may be used in multiple different combinations, and ad-hoc use is also encouraged, where one tool may provide clues and other tools executed to undertake further investigations based on initial results. 2) Automated execution of tool chains is supported by workflows. 3) support for immutable storage of audit records of tests and results is an important requirement. 4) Indicators (observations or measurements representing information of relevance for assessment of cyber security) are a key mechanism for intercommunication between one tool and another, or with the operator. 5) Mapping this work to established security development lifecycles is a useful means of determining applicability and utility of the tools and framework. 6) There is a key interplay between devices and systems. 7) Anomaly detection in multiple forms is a key means of runtime monitoring. 8) Considerable investigation is needed related to the specifics of each device / system as an item of further work., Comment: Preprint of Paper submitted to Springer Nature Computer Science

While Large Language Models (LLMs) demonstrate impressive reasoning capabilities, understanding and validating their knowledge utilization remains challenging. Chain-of-thought (CoT) prompting partially addresses this by revealing intermediate reasoning steps, but the knowledge flow and application remain implicit. We introduce IAO (Input-Action-Output) prompting, a structured template-based method that explicitly models how LLMs access and apply their knowledge during complex reasoning tasks. IAO decomposes problems into sequential steps, each clearly identifying the input knowledge being used, the action being performed, and the resulting output. This structured decomposition enables us to trace knowledge flow, verify factual consistency, and identify potential knowledge gaps or misapplications. Through experiments across diverse reasoning tasks, we demonstrate that IAO not only improves zero-shot performance but also provides transparency in how LLMs leverage their stored knowledge. Human evaluation confirms that this structured approach enhances our ability to verify knowledge utilization and detect potential hallucinations or reasoning errors. Our findings provide insights into both knowledge representation within LLMs and methods for more reliable knowledge application., Comment: Accepted as Oral at KnowFM @ AAAI 2025

The International Semantic Web Research School (ISWS) is a week-long intensive program designed to immerse participants in the field. This document reports a collaborative effort performed by ten teams of students, each guided by a senior researcher as their mentor, attending ISWS 2023. Each team provided a different perspective to the topic of creative AI, substantiated by a set of research questions as the main subject of their investigation. The 2023 edition of ISWS focuses on the intersection of Semantic Web technologies and Creative AI. ISWS 2023 explored various intersections between Semantic Web technologies and creative AI. A key area of focus was the potential of LLMs as support tools for knowledge engineering. Participants also delved into the multifaceted applications of LLMs, including legal aspects of creative content production, humans in the loop, decentralised approaches to multimodal generative AI models, nanopublications and AI for personal scientific knowledge graphs, commonsense knowledge in automatic story and narrative completion, generative AI for art critique, prompt engineering, automatic music composition, commonsense prototyping and conceptual blending, and elicitation of tacit knowledge. As Large Language Models and semantic technologies continue to evolve, new exciting prospects are emerging: a future where the boundaries between creative expression and factual knowledge become increasingly permeable and porous, leading to a world of knowledge that is both informative and inspiring., Comment: Technical Report

Efficient, low-noise, and high-bandwidth transduction between optical and microwave photons is key to long-range quantum communication between distant superconducting quantum processors. Recent demonstrations of microwave-optical transduction using the broadband direct electro-optic (Pockels) effect in optical thin films made of AlN or LiNbO$_3$ have shown promise. To improve efficiency and added noise, materials with larger Pockels coefficients, such as the soft ferroelectrics BaTiO$_3$ or SrTiO$_3$, are required. However, these materials require adapted designs and fabrication approaches due to their nonlinear and, in some cases, hysteretic electro-optic response. Here, we engineer an on-chip, triply resonant transducer comprising low-loss BaTiO$_3$-on-SiO$_2$ waveguides monolithically integrated with a superconducting microwave resonator made of Nb. We demonstrate bidirectional microwave-optical transduction and reach total off-chip efficiencies of $1\times10^{-6}$ using pulsed pumping. Our novel device concept permits in-situ poling of the ferroelectric material without introducing excess microwave loss, using a fully subtractive fabrication process with superconducting air bridges. In addition, we investigate optically induced heating, revealing fast thermalization and quasiparticle resilience of the microwave resonator. Our transducer concept and fabrication process are applicable to other materials with a large bias-induced Pockels effect and pave the way for efficient, low-power quantum interconnects., Comment: 31 pages, 30 figures

Political discourse datasets are important for gaining political insights, analyzing communication strategies or social science phenomena. Although numerous political discourse corpora exist, comprehensive, high-quality, annotated datasets are scarce. This is largely due to the substantial manual effort, multidisciplinarity, and expertise required for the nuanced annotation of rhetorical strategies and ideological contexts. In this paper, we present AgoraSpeech, a meticulously curated, high-quality dataset of 171 political speeches from six parties during the Greek national elections in 2023. The dataset includes annotations (per paragraph) for six natural language processing (NLP) tasks: text classification, topic identification, sentiment analysis, named entity recognition, polarization and populism detection. A two-step annotation was employed, starting with ChatGPT-generated annotations and followed by exhaustive human-in-the-loop validation. The dataset was initially used in a case study to provide insights during the pre-election period. However, it has general applicability by serving as a rich source of information for political and social scientists, journalists, or data scientists, while it can be used for benchmarking and fine-tuning NLP and large language models (LLMs).

While new technologies emerge, human errors always looming. Software supply chain is increasingly complex and intertwined, the security of a service has become paramount to ensuring the integrity of products, safeguarding data privacy, and maintaining operational continuity. In this work, we conducted experiments on the promising open Large Language Models (LLMs) into two main software security challenges: source code language errors and deprecated code, with a focus on their potential to replace conventional static and dynamic security scanners that rely on predefined rules and patterns. Our findings suggest that while LLMs present some unexpected results, they also encounter significant limitations, particularly in memory complexity and the management of new and unfamiliar data patterns. Despite these challenges, the proactive application of LLMs, coupled with extensive security databases and continuous updates, holds the potential to fortify Software Supply Chain (SSC) processes against emerging threats.

This study introduces DK-PRACTICE (Dynamic Knowledge Prediction and Educational Content Recommendation System), an intelligent online platform that leverages machine learning to provide personalized learning recommendations based on student knowledge state. Students participate in a short, adaptive assessment using the question-and-answer method regarding key concepts in a specific knowledge domain. The system dynamically selects the next question for each student based on the correctness and accuracy of their previous answers. After the test is completed, DK-PRACTICE analyzes students' interaction history to recommend learning materials to empower the student's knowledge state in identified knowledge gaps. Both question selection and learning material recommendations are based on machine learning models trained using anonymized data from a real learning environment. To provide self-assessment and monitor learning progress, DK-PRACTICE allows students to take two tests: one pre-teaching and one post-teaching. After each test, a report is generated with detailed results. In addition, the platform offers functions to visualize learning progress based on recorded test statistics. DK-PRACTICE promotes adaptive and personalized learning by empowering students with self-assessment capabilities and providing instructors with valuable information about students' knowledge levels. DK-PRACTICE can be extended to various educational environments and knowledge domains, provided the necessary data is available according to the educational topics. A subsequent paper will present the methodology for the experimental application and evaluation of the platform., Comment: 13 pages, The Barcelona Conference on Education 2024

The expressive power of message-passing graph neural networks (MPNNs) is reasonably well understood, primarily through combinatorial techniques from graph isomorphism testing. However, MPNNs' generalization abilities -- making meaningful predictions beyond the training set -- remain less explored. Current generalization analyses often overlook graph structure, limit the focus to specific aggregation functions, and assume the impractical, hard-to-optimize $0$-$1$ loss function. Here, we extend recent advances in graph similarity theory to assess the influence of graph structure, aggregation, and loss functions on MPNNs' generalization abilities. Our empirical study supports our theoretical insights, improving our understanding of MPNNs' generalization properties.

Hate speech online remains an understudied issue for marginalized communities, and has seen rising relevance, especially in the Global South, which includes developing societies with increasing internet penetration. In this paper, we aim to provide marginalized communities living in societies where the dominant language is low-resource with a privacy-preserving tool to protect themselves from hate speech on the internet by filtering offensive content in their native languages. Our contribution in this paper is twofold: 1) we release REACT (REsponsive hate speech datasets Across ConTexts), a collection of high-quality, culture-specific hate speech detection datasets comprising seven distinct target groups in eight low-resource languages, curated by experienced data collectors; 2) we propose a solution to few-shot hate speech detection utilizing federated learning (FL), a privacy-preserving and collaborative learning approach, to continuously improve a central model that exhibits robustness when tackling different target groups and languages. By keeping the training local to the users' devices, we ensure the privacy of the users' data while benefitting from the efficiency of federated learning. Furthermore, we personalize client models to target-specific training data and evaluate their performance. Our results indicate the effectiveness of FL across different target groups, whereas the benefits of personalization on few-shot learning are not clear.

Given points from an arbitrary metric space and a sequence of point updates sent by an adversary, what is the minimum recourse per update (i.e., the minimum number of changes needed to the set of centers after an update), in order to maintain a constant-factor approximation to a $k$-clustering problem? This question has received attention in recent years under the name consistent clustering. Previous works by Lattanzi and Vassilvitskii [ICLM '17] and Fichtenberger, Lattanzi, Norouzi-Fard, and Svensson [SODA '21] studied $k$-clustering objectives, including the $k$-center and the $k$-median objectives, under only point insertions. In this paper we study the $k$-center objective in the fully dynamic setting, where the update is either a point insertion or a point deletion. Before our work, {\L}\k{a}cki, Haeupler, Grunau, Rozho\v{n}, and Jayaram [SODA '24] gave a deterministic fully dynamic constant-factor approximation algorithm for the $k$-center objective with worst-case recourse of $2$ per update. In this work, we prove that the $k$-center clustering problem admits optimal recourse bounds by developing a deterministic fully dynamic constant-factor approximation algorithm with worst-case recourse of $1$ per update. Moreover our algorithm performs simple choices based on light data structures, and thus is arguably more direct and faster than the previous one which uses a sophisticated combinatorial structure. Additionally, we develop a new deterministic decremental algorithm and a new deterministic incremental algorithm, both of which maintain a $6$-approximate $k$-center solution with worst-case recourse of $1$ per update. Our incremental algorithm improves over the $8$-approximation algorithm by Charikar, Chekuri, Feder, and Motwani [STOC '97]. Finally, we remark that since all three of our algorithms are deterministic, they work against an adaptive adversary., Comment: In Proceedings SODA 2025

This paper studies stabilization of linear time-invariant (LTI) systems when control actions can only be realized in finitely many directions where it is possible to actuate uniformly or logarithmically extended positive scaling factors in each direction. Furthermore, a nearest-action selection approach is used to map the continuous measurements to a realizable action where we show that the approach satisfies a weak sector condition for multiple-input multiple-output (MIMO) systems. Using the notion of input-to-state stability, under some assumptions imposed on the transfer function of the system, we show that the closed-loop system converges to the target ball exponentially fast. Moreover, when logarithmic extension for the scaling factors is realizable, the closed-loop system is able to achieve asymptotic stability instead of only practical stability. Finally, we present an example of the application that confirms our analysis., Comment: 10 pages, 5 figures

Traditional studies of memory for meaningful narratives focus on specific stories and their semantic structures but do not address common quantitative features of recall across different narratives. We introduce a statistical ensemble of random trees to represent narratives as hierarchies of key points, where each node is a compressed representation of its descendant leaves, which are the original narrative segments. Recall is modeled as constrained by working memory capacity from this hierarchical structure. Our analytical solution aligns with observations from large-scale narrative recall experiments. Specifically, our model explains that (1) average recall length increases sublinearly with narrative length, and (2) individuals summarize increasingly longer narrative segments in each recall sentence. Additionally, the theory predicts that for sufficiently long narratives, a universal, scale-invariant limit emerges, where the fraction of a narrative summarized by a single recall sentence follows a distribution independent of narrative length., Comment: 21 pages, 5 figures; included new derivations

In this research, wettability control by area fraction of laser ablated surface of copper is presented. The functional surfaces with full wettability control from highly hydrophilic to super-hydrophobic were created on copper by nanosecond (ns) and picosecond (ps) laser irradiation. The area fraction and color change were evaluated by digital image processing of microscopic images of the laser-ablated copper surface. The control of the wetting angle from almost 0 degrees to 132 degrees was achieved for both ps and ns pulses by controlling the area fraction of the laser-ablated surface. Cassie, Cassie-Baxter, and Wenzel models were adopted to explain the experimental results. For the first time, the wettability and color of copper were controlled by controlling the area fraction of the laser-ablated surface. It is expected that the current results make an impact on the heat exchanger technology of water heat sinks, cooling units, atmospheric water generators, and fog harvesting and impact numerous applications from power plants to solar thermal water systems devices where highly-hydrophilic to super hydrophobic copper can be applied., Comment: 30 pages, 11 figures, 1 table, 103 references

We present Y-MAP-Net, a Y-shaped neural network architecture designed for real-time multi-task learning on RGB images. Y-MAP-Net, simultaneously predicts depth, surface normals, human pose, semantic segmentation and generates multi-label captions, all from a single network evaluation. To achieve this, we adopt a multi-teacher, single-student training paradigm, where task-specific foundation models supervise the network's learning, enabling it to distill their capabilities into a lightweight architecture suitable for real-time applications. Y-MAP-Net, exhibits strong generalization, simplicity and computational efficiency, making it ideal for robotics and other practical scenarios. To support future research, we will release our code publicly., Comment: 8 page paper, 6 Figures, 3 Tables

Synthetic biology is a recent area of biological engineering, whose aim is to provide cells with novel functionalities. A number of important results regarding the development of control circuits in synthetic biology have been achieved during the last decade. A differential geometry approach can be used for the analysis of said systems, which are often nonlinear. Here we demonstrate the application of such tools to analyse the structural identifiability, observability, accessibility, and controllability of several biomolecular systems. We focus on a set of synthetic circuits of current interest, which can perform several tasks, both in open loop and closed loop settings. We analyse their properties with our own methods and tools; further, we describe a new open-source implementation of the techniques., Comment: 7 pages, 1 figure, submitted to DYCOPS2025

This volume contains the proceedings of GandALF 2024, the Fifteenth International Symposium on Games, Automata, Logics, and Formal Verification. GandALF 2024 took place on 19-21 June 2024, in Reykjavik, Iceland. The aim of GandALF 2024 is to bring together researchers from academia and industry who are actively working in the fields of Games, Automata, Logics, and Formal Verification. The idea is to cover an ample spectrum of themes, ranging from theory to applications, and stimulate cross-fertilization.

The sequential execution of actions and their hierarchical structure consisting of different levels of abstraction, provide features that remain unexplored in the task of action recognition. In this study, we present a novel approach to improve action recognition by exploiting the hierarchical organization of actions and by incorporating contextualized textual information, including location and prior actions to reflect the sequential context. To achieve this goal, we introduce a novel transformer architecture tailored for action recognition that utilizes both visual and textual features. Visual features are obtained from RGB and optical flow data, while text embeddings represent contextual information. Furthermore, we define a joint loss function to simultaneously train the model for both coarse and fine-grained action recognition, thereby exploiting the hierarchical nature of actions. To demonstrate the effectiveness of our method, we extend the Toyota Smarthome Untrimmed (TSU) dataset to introduce action hierarchies, introducing the Hierarchical TSU dataset. We also conduct an ablation study to assess the impact of different methods for integrating contextual and hierarchical data on action recognition performance. Results show that the proposed approach outperforms pre-trained SOTA methods when trained with the same hyperparameters. Moreover, they also show a 17.12% improvement in top-1 accuracy over the equivalent fine-grained RGB version when using ground-truth contextual information, and a 5.33% improvement when contextual information is obtained from actual predictions.

Software engineers operating in complex and dynamic environments must continuously adapt to evolving requirements, learn iteratively from experience, and reconsider their approaches based on new insights. However, current large language model (LLM)-based software agents often follow linear, sequential processes that prevent backtracking and exploration of alternative solutions, limiting their ability to rethink their strategies when initial approaches prove ineffective. To address these challenges, we propose SWE-Search, a multi-agent framework that integrates Monte Carlo Tree Search (MCTS) with a self-improvement mechanism to enhance software agents' performance on repository-level software tasks. SWE-Search extends traditional MCTS by incorporating a hybrid value function that leverages LLMs for both numerical value estimation and qualitative evaluation. This enables self-feedback loops where agents iteratively refine their strategies based on both quantitative numerical evaluations and qualitative natural language assessments of pursued trajectories. The framework includes a SWE-Agent for adaptive exploration, a Value Agent for iterative feedback, and a Discriminator Agent that facilitates multi-agent debate for collaborative decision-making. Applied to the SWE-bench benchmark, our approach demonstrates a 23% relative improvement in performance across five models compared to standard open-source agents without MCTS. Our analysis reveals how performance scales with increased inference-time compute through deeper search, providing a pathway to improve software agents without requiring larger models or additional training data. This highlights the potential of self-evaluation driven search techniques in complex software engineering environments., Comment: Main body: 10 pages, 5 figures. Appendix: 5 pages, 4 figures. Open-source codebase

The features of self-supervised vision transformers (ViTs) contain strong semantic and positional information relevant to downstream tasks like object localization and segmentation. Recent works combine these features with traditional methods like clustering, graph partitioning or region correlations to achieve impressive baselines without finetuning or training additional networks. We leverage upsampled features from ViT networks (e.g DINOv2) in two workflows: in a clustering based approach for object localization and segmentation, and paired with standard classifiers in weakly supervised materials segmentation. Both show strong performance on benchmarks, especially in weakly supervised segmentation where the ViT features capture complex relationships inaccessible to classical approaches. We expect the flexibility and generalizability of these features will both speed up and strengthen materials characterization, from segmentation to property-prediction.

This chapter introduces a research project titled "Analyzing the Political Discourse: A Collaboration Between Humans and Artificial Intelligence", which was initiated in preparation for Greece's 2023 general elections. The project focused on the analysis of political leaders' campaign speeches, employing Artificial Intelligence (AI), in conjunction with an interdisciplinary team comprising journalists, a political scientist, and data scientists. The chapter delves into various aspects of political discourse analysis, including sentiment analysis, polarization, populism, topic detection, and Named Entities Recognition (NER). This experimental study investigates the capabilities of large language model (LLMs), and in particular OpenAI's ChatGPT, for analyzing political speech, evaluates its strengths and weaknesses, and highlights the essential role of human oversight in using AI in journalism projects and potentially other societal sectors. The project stands as an innovative example of human-AI collaboration (known also as "hybrid intelligence") within the realm of digital humanities, offering valuable insights for future initiatives.

We present D-PoSE (Depth as an Intermediate Representation for 3D Human Pose and Shape Estimation), a one-stage method that estimates human pose and SMPL-X shape parameters from a single RGB image. Recent works use larger models with transformer backbones and decoders to improve the accuracy in human pose and shape (HPS) benchmarks. D-PoSE proposes a vision based approach that uses the estimated human depth-maps as an intermediate representation for HPS and leverages training with synthetic data and the ground-truth depth-maps provided with them for depth supervision during training. Although trained on synthetic datasets, D-PoSE achieves state-of-the-art performance on the real-world benchmark datasets, EMDB and 3DPW. Despite its simple lightweight design and the CNN backbone, it outperforms ViT-based models that have a number of parameters that is larger by almost an order of magnitude. D-PoSE code is available at: https://github.com/nvasilik/D-PoSE

We examine how skewness interacts with kurtosis within the cryptocurrency market. We show that during the COVID-19 pandemic there are more clusters of observations around the two flanks, highlighting the presence of a volatile behavior. Moreover, we document the evolvement of the interrelationship as the pandemic progresses, identifying the domination of the extremes. Our findings advance the thinking that by exploiting the interrelationship between the two higher moments of cryptocurrencies, investors and researchers can have in their arsenal an additional analytic tool.

Fraud is a prevalent offence that extends beyond financial loss, causing psychological and physical harm to victims. The advancements in online communication technologies alowed for online fraud to thrive in this vast network, with fraudsters increasingly using these channels for deception. With the progression of technologies like AI, there is a growing concern that fraud will scale up, using sophisticated methods, like deep-fakes in phishing campaigns, all generated by language generation models like ChatGPT. However, the application of AI in detecting and analyzing online fraud remains understudied. We conduct a Systematic Literature Review on AI and NLP techniques for online fraud detection. The review adhered the PRISMA-ScR protocol, with eligibility criteria including relevance to online fraud, use of text data, and AI methodologies. We screened 2,457 academic records, 350 met our eligibility criteria, and included 223. We report the state-of-the-art NLP techniques for analysing various online fraud categories; the training data sources; the NLP algorithms and models built; and the performance metrics employed for model evaluation. We find that current research on online fraud is divided into various scam activitiesand identify 16 different frauds that researchers focus on. This SLR enhances the academic understanding of AI-based detection methods for online fraud and offers insights for policymakers, law enforcement, and businesses on safeguarding against such activities. We conclude that focusing on specific scams lacks generalization, as multiple models are required for different fraud types. The evolving nature of scams limits the effectiveness of models trained on outdated data. We also identify issues in data limitations, training bias reporting, and selective presentation of metrics in model performance reporting, which can lead to potential biases in model evaluation., Comment: Manuscript under peer review. Content may be subject to revision in future versions

In computer graphics (CG) education, the challenge of finding modern, versatile tools is significant, particularly when integrating both legacy and advanced technologies. Traditional frameworks, often reliant on solid, yet outdated APIs like OpenGL, limit the exploration of cutting-edge graphics techniques. To address this, we introduce pyGANDALF, a unique, lightweight, open-source CG framework built on three pillars: Entity-Component-System (ECS) architecture, Python programming, and WebGPU integration. This combination sets pyGANDALF apart by providing a streamlined ECS design with an editor layer, compatibility with WebGPU for state-of-the-art features like compute and ray tracing pipelines, and a programmer-friendly Python environment. The framework supports modern features, such as Physically Based Rendering (PBR) capabilities and integration with Universal Scene Description (USD) formats, making it suitable for both educational demonstrations and real-world applications. Evaluations by expert users confirmed that pyGANDALF effectively balances ease of use with advanced functionality, preparing students for contemporary CG development challenges., Comment: 8 pages, 6 Figures

Given that AI systems are set to play a pivotal role in future decision-making processes, their trustworthiness and reliability are of critical concern. Due to their scale and complexity, modern AI systems resist direct interpretation, and alternative ways are needed to establish trust in those systems, and determine how well they align with human values. We argue that good measures of the information processing similarities between AI and humans, may be able to achieve these same ends. While Representational alignment (RA) approaches measure similarity between the internal states of two systems, the associated data can be expensive and difficult to collect for human systems. In contrast, Behavioural alignment (BA) comparisons are cheaper and easier, but questions remain as to their sensitivity and reliability. We propose two new behavioural alignment metrics misclassification agreement which measures the similarity between the errors of two systems on the same instances, and class-level error similarity which measures the similarity between the error distributions of two systems. We show that our metrics correlate well with RA metrics, and provide complementary information to another BA metric, within a range of domains, and set the scene for a new approach to value alignment.

Transformers demonstrate competitive performance in terms of precision on the problem of vision-based object detection. However, they require considerable computational resources due to the quadratic size of the attention weights. In this work, we propose to cluster the transformer input on the basis of its entropy. The reason for this is that the self-information of each pixel (whose sum is the entropy), is likely to be similar among pixels corresponding to the same objects. Clustering reduces the size of data given as input to the transformer and therefore reduces training time and GPU memory usage, while at the same time preserves meaningful information to be passed through the remaining parts of the network. The proposed process is organized in a module called ENACT, that can be plugged-in any transformer architecture that consists of a multi-head self-attention computation in its encoder. We ran extensive experiments using the COCO object detection dataset, and three detection transformers. The obtained results demonstrate that in all tested cases, there is consistent reduction in the required computational resources, while the precision of the detection task is only slightly reduced. The code of the ENACT module will become available at https://github.com/GSavathrakis/ENACT

Control barrier functions (CBFs) offer an efficient framework for designing real-time safe controllers. However, CBF-based controllers can be short-sighted, resulting in poor performance, a behaviour which is aggravated in uncertain conditions. This motivated research on safety filters based on model predictive control (MPC) and its stochastic variant. MPC deals with safety constraints in a direct manner, however, its computational demands grow with the prediction horizon length. We propose a safety formulation that solves a finite horizon optimization problem at each time instance like MPC, but rather than explicitly imposing constraints along the prediction horizon, we enforce probabilistic safety constraints by means of CBFs only at the first step of the horizon. The probabilistic CBF constraints are transformed in a finite number of deterministic CBF constraints via the scenario based methodology. Capitalizing on results on scenario based MPC, we provide distribution-free, \emph{a priori} guarantees on the system's closed loop expected safety violation frequency. We demonstrate our results through a case study on unmanned aerial vehicle collision-free position swapping, and provide a numerical comparison with recent stochastic CBF formulations., Comment: To be published in: The 63rd IEEE Conference on Decision and Control (CDC-2024 Milano, Italy)

Machine Learning (ML) has become one of the most impactful fields of data science in recent years. However, a significant concern with ML is its privacy risks due to rising attacks against ML models. Privacy-Preserving Machine Learning (PPML) methods have been proposed to mitigate the privacy and security risks of ML models. A popular approach to achieving PPML uses Homomorphic Encryption (HE). However, the highly publicized inefficiencies of HE make it unsuitable for highly scalable scenarios with resource-constrained devices. Hence, Hybrid Homomorphic Encryption (HHE) -- a modern encryption scheme that combines symmetric cryptography with HE -- has recently been introduced to overcome these challenges. HHE potentially provides a foundation to build new efficient and privacy-preserving services that transfer expensive HE operations to the cloud. This work introduces HHE to the ML field by proposing resource-friendly PPML protocols for edge devices. More precisely, we utilize HHE as the primary building block of our PPML protocols. We assess the performance of our protocols by first extensively evaluating each party's communication and computational cost on a dummy dataset and show the efficiency of our protocols by comparing them with similar protocols implemented using plain BFV. Subsequently, we demonstrate the real-world applicability of our construction by building an actual PPML application that uses HHE as its foundation to classify heart disease based on sensitive ECG data., Comment: Accepted in The 22nd IEEE International Conference on Dependable, Autonomic and Secure Computing (DASC 2024)

Supermassive Black Holes (BHs) are known to efficiently grow through gas accretion, but even sustained and intense mass build-up through this mechanism struggles to explain the assembly of the most massive BHs observed in the local Universe. Using the Chandra Deep-Wide Field Survey (CDFWS) in the Bo\"otes field, we measure BH--galaxy assembly in massive galaxies ($M_\star\gtrsim10^{10}\rm M_\odot$) through the AGN fraction and specific Black Hole accretion rate (sBHAR) distribution as a function of redshift and stellar mass. We determine stellar masses and star formation rates for a parent sample of optically selected galaxies as well as those with X-ray detections indicating the presence of an AGN through Spectral Energy Distribution (SED) fitting. We derive a redshift-dependent mass completeness limit and extract X-ray information for every galaxy as to provide a comprehensive picture of the AGN population in massive galaxies. While X-ray AGN samples are dominated by moderately massive host galaxies of $M_{\star} \geqslant 10^{10}\rm M_{\odot}$, we do not find a strong stellar mass dependence in AGN fraction (to limits in sBHAR), indicating a bias towards massive galaxies in the observed samples. We derive BH-galaxy growth tracks over time, which reveal that while most BH mass has been accumulated since $z=4$ for lower mass BHs, the assembly of the most massive BHs is more complex, with little to no relative mass gain since $z=4$, implying that rapid and intense growth episodes prior to $z=4$ were necessary to form these massive BHs., Comment: 21 pages, 13 figures, accepted to MNRAS

We consider backward stochastic differential equations (BSDEs) with mean-field and McKean-Vlasov interactions in their generators in a general setting, where the drivers are square-integrable martingales, with a focus on the independent increments case, and the filtrations are (possibly) stochastically discontinuous. In other words, we consider discrete- and continuous-time systems of mean-field BSDEs and McKean-Vlasov BSDEs in a unified setting. We provide existence and uniqueness results for these BSDEs using new a priori estimates that utilize the stochastic exponential. Then, we provide propagation of chaos results for systems of particles that satisfy BSDEs, i.e. we show that the asymptotic behaviour of the solutions of mean-field systems of BSDEs, as the multitude of the systems grows to infinity, converges to I.I.D solutions of McKean-Vlasov BSDEs. We introduce a new technique for showing the backward propagation of chaos, that makes repeated use of the a priori estimates, inequalities for the Wasserstein distance and the ``conservation of solutions'' under different filtrations, and does not demand the solutions of the mean field systems to be exchangeable or symmetric. Finally, we deduce convergence rates for the propagation of chaos, under advanced integrability conditions on the solutions of the BSDEs.

Large language models (LLMs) are poised to revolutionize the domain of online fashion retail, enhancing customer experience and discovery of fashion online. LLM-powered conversational agents introduce a new way of discovery by directly interacting with customers, enabling them to express in their own ways, refine their needs, obtain fashion and shopping advice that is relevant to their taste and intent. For many tasks in e-commerce, such as finding a specific product, conversational agents need to convert their interactions with a customer to a specific call to different backend systems, e.g., a search system to showcase a relevant set of products. Therefore, evaluating the capabilities of LLMs to perform those tasks related to calling other services is vital. However, those evaluations are generally complex, due to the lack of relevant and high quality datasets, and do not align seamlessly with business needs, amongst others. To this end, we created a multilingual evaluation dataset of 4k conversations between customers and a fashion assistant in a large e-commerce fashion platform to measure the capabilities of LLMs to serve as an assistant between customers and a backend engine. We evaluate a range of models, showcasing how our dataset scales to business needs and facilitates iterative development of tools., Comment: Accepted at KDD workshop on Evaluation and Trustworthiness of Generative AI Models

A Hereditarily Indecomposable (HI) Banach space $X$ admits an HI extension if there exists an HI space $Z$ such that $X$ is isomorphic to a subspace $Y$ of $Z$ and $Z/Y$ is of infinite dimension. The problem whether or not every HI space admits an HI extension is attributed to A. Pelczynski. In this paper we present a method to define HI-extensions of the standard HI spaces, a class which includes the Gowers-Maurey space, asymptotic $\ell_{p}$-HI spaces and others., Comment: 27 pages, no figures

The impressive capabilities of large language models (LLMs) have sparked debate over whether these models genuinely generalize to unseen tasks or predominantly rely on memorizing vast amounts of pretraining data. To explore this issue, we introduce an extended concept of memorization, distributional memorization, which measures the correlation between the LLM output probabilities and the pretraining data frequency. To effectively capture task-specific pretraining data frequency, we propose a novel task-gram language model, which is built by counting the co-occurrence of semantically related $n$-gram pairs from task inputs and outputs in the pretraining corpus. Using the Pythia models trained on the Pile dataset, we evaluate four distinct tasks: machine translation, factual question answering, world knowledge understanding, and math reasoning. Our findings reveal varying levels of memorization, with the strongest effect observed in factual question answering. Furthermore, while model performance improves across all tasks as LLM size increases, only factual question answering shows an increase in memorization, whereas machine translation and reasoning tasks exhibit greater generalization, producing more novel outputs. This study demonstrates that memorization plays a larger role in simpler, knowledge-intensive tasks, while generalization is the key for harder, reasoning-based tasks, providing a scalable method for analyzing large pretraining corpora in greater depth., Comment: Accepted to ICLR 2025

This work proposes a novel approach to transform any modern game engine pipeline, for optimized performance and enhanced user experiences in Extended Reality (XR) environments. Decoupling the physics engine from the game engine pipeline and using a client-server N-1 architecture creates a scalable solution, efficiently serving multiple graphics clients on Head-Mounted Displays (HMDs) with a single physics engine on edge-cloud infrastructure. This approach ensures better synchronization in multiplayer scenarios without introducing overhead in single-player experiences, maintaining session continuity despite changes in user participation. Relocating the Physics Engine to an edge or cloud node reduces strain on local hardware, dedicating more resources to high-quality rendering and unlocking the full potential of untethered HMDs. We present four algorithms that decouple the physics engine, increasing frame rates and Quality of Experience (QoE) in VR simulations, supporting advanced interactions, numerous physics objects, and multi-user sessions with over 100 concurrent users. Incorporating a Geometric Algebra interpolator reduces inter-calls between dissected parts, maintaining QoE and easing network stress. Experimental validation, with more than 100 concurrent users, 10,000 physics objects, and softbody simulations, confirms the technical viability of the proposed architecture, showcasing transformative capabilities for more immersive and collaborative XR applications without compromising performance., Comment: 11 pages, 11 figures, 1 table, 4 algorithms, CGI 2024, Computer Animation and Virtual Worlds

Automated Cryptographic Validation Protocol (ACVP) is an existing protocol that is used to validate a software or hardware cryptographic module automatically. In this work, we present a system providing the method and tools to produce well-covering tests in ACVP format for cryptographic libraries. The system achieves better coverage than existing fuzzing methods by using a hybrid approach to fuzzing cryptographic primitives. In addition, the system offers a framework that allows to creates easily and securely create testing modules for cryptographic libraries. The work demonstrates how this system has been used to improve automated testing of NSS (Network Security Services), a popular cryptographic library, detect its vulnerabilities and suggest ways to improve and further develop the ACVP test format.