Archivio della ricerca - Fondazione Bruno Kessler
Not a member yet
21227 research outputs found
Sort by
STELVIO: Exploring Factor Graphs for a Robust Stereo-Visual-LiDAR-Inertial Odometry
No full textAccurate and robust odometry is critical for mobile mapping and autonomous navigation, particularly in complex environments where single-sensor approaches struggle. While LiDAR and visual odometry each provide valuable motion estimation, they are susceptible to failures in conditions unfavorable for the specific odometry type. Fusing multiple modalities enhances robustness, yet effective integration remains challenging due to differences in heterogenous sensor data representation. This study presents STELVIO, a flexible factor graph-based framework for Stereo LiDAR-Visual-Inertial Odometry. By combining LiDAR-inertial odometry with stereo visual odometry, STELVIO improves trajectory estimation by leveraging the strengths of each modality. The system introduces adaptive fusion strategies, ranging from loose, odometry-only pose graph coupling to an extensive factor graph approach, utilizing visual features and LiDAR-derived range factors. This modular structure allows for balancing computational efficiency with robustness, making it suitable for real-time applications and accuracy-oriented mapping applications. Evaluation is conducted using an in-house mobile mapping system in a challenging indoor environment. Initial results highlight the effectiveness of the fusion approach in reducing drift and improving localization consistency compared to single-sensor methods. The findings demonstrate the potential of multi-sensor integration for robust and scalable mobile mapping solutions
You don't bring me flowers: Mitigating unwanted recommendations through conformal risk control
No full textRecommenders are significantly shaping online information consumption. While effective at personalizing content, these systems increasingly face criticism for propagating irrelevant, unwanted, and even harmful recommendations. Such content degrades user satisfaction and contributes to significant societal issues, including misinformation, radicalization, and erosion of user trust. Although platforms offer mechanisms to mitigate exposure to undesired content, these mechanisms are often insufficiently effective and slow to adapt to users’ feedback. This paper introduces an intuitive, model-agnostic, and distribution-free method that uses conformal risk control to provably bound unwanted content in personalized recommendations by leveraging simple binary feedback on items. We also address a limitation of traditional conformal risk control approaches, i.e., the fact that the recommender can provide a smaller set of recommended items, by leveraging implicit feedback on consumed items to expand the recommendation set while ensuring robust risk mitigation. Our experimental evaluation on data coming from a popular online video-sharing platform demonstrates that our approach ensures an effective and controllable reduction of unwanted recommendations with minimal effort. The source code is available here: https://github.com/geektoni/mitigating-harm-recsys
Thin LGADs and thin silicon diodes for applications in radiotherapy
No full textLow gain avalanche diodes (LGADs) and thin n-on-p silicon diodes, when read out by fast and custom electronics, exhibit characteristics that make them promising candidates for the development of new detectors for clinical applications such as beam commissioning, diagnostics and monitoring, dosimetry, and online treatment delivery verification. Compared to gas ionization chambers, these detectors offer significantly higher sensitivity, enabling the detection of single particles at fluxes of up to 10
8
particles/cm
2
s—sufficient to cover the entire clinical intensity range of carbon ion therapy and approximately one order of magnitude lower for proton therapy. Various front-end electronics have been developed and characterized for readout configurations, ranging from single channels (pads or strips) to arrays of up to 144 strips. These systems have been applied to single-particle identification for beam monitors in particle therapy, as well as to two-dimensional beam monitoring and dosimetry in ultra-high dose rate and spatially fractionated radiotherapy. This review summarizes the detectors based on LGADs and thin n-on-p silicon diodes developed within the INFN-CSN5 projects MoVeIT, SIG, and FRIDA. Specifically, we present a 2.7 × 2.7 cm
2
particle counter for measuring beam fluence and position, a beam energy detector based on the primary particle’s time-of-flight, a setup for studying beam time structure at the nanosecond scale, and a system for range verification
via
prompt gamma timing. Current advances in various technologies are reviewed, together with challenges and future perspectives on the application of LGADs and thin silicon diodes in radiotherapy
New measurements of light yield quenching in EJ-200 and LYSO scintillators
No full textLutetium–Yttrium OxyorthoSilicate (LYSO) crystals and EJ-200 plastic are widely recognized fast scintillators, valued for their high light output and robust mechanical properties, which make them well-suited for high-energy physics and space applications. The non-proportional light response of these scintillators, alongside their non-linear behavior at low-energy X-rays, has been extensively studied. Nevertheless, discrepancies in published measurements of light yield quenching remain unresolved. This study provides new insights by measuring light quenching in both scintillators under charged particle excitation and analyzing the non-linear response of LYSO to X-ray and
-ray sources. The results are interpreted using the modified Birks–Onsager model, offering a reasonable description for the observed scintillation quenching phenomena
On the Energy Consumption of Test Generation
No full textResearch in the area of automated test generation has seen remarkable progress in recent years, resulting in several approaches and tools for effective and efficient generation of test cases. In particular, the EvoSuite tool has been at the forefront of this progress embodying various algorithms for automated test generation of Java programs. EvoSuite has been used to generate test cases for a wide variety of programs as well. While there are a number of empirical studies that report results on the effectiveness, in terms of code coverage and other related metrics, of the various test generation strategies and algorithms implemented in EvoSuite, there are no studies, to the best of our knowledge, on the energy consumption associated to the automated test generation. In this paper, we set out to investigate this aspect by measuring the energy consumed by EvoSuite when generating tests. We also measure the energy consumed in the execution of the test cases generated, comparing them with those manually written by developers. The results show that the different test generation algorithms consumed different amounts of energy, in particular on classes with high cyclomatic complexity. Furthermore, we also observe that manual tests tend to consume more energy as compared to automatically generated tests, without necessarily achieving higher code coverage. Our results also give insight into the methods that consume significantly higher levels of energy, indicating potential points of improvement both for EvoSuite as well as the different programs under test
Leveraging a Hybrid Quantum-Classical Framework for Subsurface Target Detection in Radar Sounding System: Challenges and Opportunities
No full textIn this article, we explore the potential of quantum machine learning (QML) for subsurface feature extractions from radar sounder (RS) signals. We propose a hybrid quantum-classical (HQC) learning paradigm that leverages parameterized quantum circuits (PQCs) to generate probability amplitudes based on quantum properties such as superposition and entanglement. These amplitudes are synergistically integrated with the classical deep neural networks that are efficient in learning high-dimensional contextual features for downstream prediction tasks. The present research work is structured around two objectives. First, we investigate the role of quantum circuits in the latent space for transferring back-and-forth rich discriminative spatial context from the encoder to the decoder for segmentation. Second, we investigate how the probabilistic amplitudes derived from quantum circuits are significant in integrating into the classical models to provide new insights for RS signals segmentation. The performance of the hybrid architectures has been studied in small-scale settings by simulating the expected behavior of the quantum circuits on a classical machine. The experimental results have demonstrated the viability of QML frameworks on MCoRDS-1 and MCoRDS-3 datasets for RS signal segmentation. Qualitatively, they are capable of delineating the spatial extent of the bedrock from noise. Additionally, we conduct a comparative analysis between the Qiskit Aer Simulator and the IBM FakeBackend Simulator to highlight the computational trade-offs and validate fidelity of two simulators for scalable experimentation. Therefore, our work opens up new avenues of research for future RS data analysis leading to more precise and efficient subsurface target segmentation
Multilingual Analysis of Narrative Properties in Conspiracist vs Mainstream Telegram Channels
No full textConspiracist narratives posit an omnipotent, evil group causing harm throughout domains. However, modern-day online conspiracism is often more erratic, consisting of loosely connected posts displaying a general anti-establishment attitude pervaded by negative emotions. We gather a dataset of 300 conspiracist and mainstream, Telegram channels in Italian and English and use the automatic extraction of entities and emotion detection to compare structural characteristics of both types of channels. We create a co-occurrence network of entities to analyze how the different types of channels introduce and use them across posts and topics. We find that conspiracist channels are characterized by anger. Moreover, co-occurrence networks of entities appearing in conspiracist channels are more dense. We theorize that this reflects a narrative structure where all actants are pushed into a single domain. Conspiracist channels disproportionately associate the most central group of entities with anger and fear. We do not find evidence that entities in conspiracist narratives occur across more topics. This could indicate an erratic type of online conspiracism where everything can be connected to everything and that is characterized by a high number of entities and high levels of anger
Evolv-1 at the ICST 2025 Tool Competition – UAV Testing Track
No full textEvolv-1 is a test case generation tool designed using Evolutionary Algorithms (EAs) to optimize UAV testing scenarios. This short paper presents Evolv-1's implementation as part of the ICST 2025 UAV Testing Tool Competition
