Nicolò Oreste Pinciroli Vago

Riccardo Campi, Nicolò Oreste Pinciroli Vago, Mathyas Giudici et al.

Retrieval-Augmented Generation (RAG) over Knowledge Graphs (KGs) suffers from the fact that indexing approaches may lose important contextual nuance when text is reduced to triples, thereby degrading performance in downstream Question-Answering (QA) tasks, particularly for multi-hop QA, which requires composing answers from multiple entities, facts, or relations. We propose a domain-agnostic, KG-based QA framework that covers both the indexing and retrieval/inference phases. A new indexing approach called Map-Disambiguate-Enrich-Reduce (MDER) generates context-derived triple descriptions and subsequently integrates them with entity-level summaries, thus avoiding the need for explicit traversal of edges in the graph during the QA retrieval phase. Complementing this, we introduce Decompose-Resolve (DR), a retrieval mechanism that decomposes user queries into resolvable triples and grounds them in the KG via iterative reasoning. Together, MDER and DR form an LLM-driven QA pipeline that is robust to sparse, incomplete, and complex relational data. Experiments show that on standard and domain specific benchmarks, MDER-DR achieves substantial improvements over standard RAG baselines (up to 66%), while maintaining cross-lingual robustness. Our code is available at https://github.com/DataSciencePolimi/MDER-DR_RAG.

Nicolò Oreste Pinciroli Vago, Piero Fraternali

Gravitational lensing is the relativistic effect generated by massive bodies, which bend the space-time surrounding them. It is a deeply investigated topic in astrophysics and allows validating theoretical relativistic results and studying faint astrophysical objects that would not be visible otherwise. In recent years Machine Learning methods have been applied to support the analysis of the gravitational lensing phenomena by detecting lensing effects in data sets consisting of images associated with brightness variation time series. However, the state-of-art approaches either consider only images and neglect time-series data or achieve relatively low accuracy on the most difficult data sets. This paper introduces DeepGraviLens, a novel multi-modal network that classifies spatio-temporal data belonging to one non-lensed system type and three lensed system types. It surpasses the current state of the art accuracy results by $\approx 3\%$ to $\approx 11\%$, depending on the considered data set. Such an improvement will enable the acceleration of the analysis of lensed objects in upcoming astrophysical surveys, which will exploit the petabytes of data collected, e.g., from the Vera C. Rubin Observatory.

Riccardo Campi, Nicolò Oreste Pinciroli Vago, Mathyas Giudici et al.

In this work, we investigate the use of Large Language Models (LLMs) within a graph-based Retrieval Augmented Generation (RAG) architecture for Energy Efficiency (EE) Question Answering. First, the system automatically extracts a Knowledge Graph (KG) from guidance and regulatory documents in the energy field. Then, the generated graph is navigated and reasoned upon to provide users with accurate answers in multiple languages. We implement a human-based validation using the RAGAs framework properties, a validation dataset comprising 101 question-answer pairs, and domain experts. Results confirm the potential of this architecture and identify its strengths and weaknesses. Validation results show how the system correctly answers in about three out of four of the cases (75.2 +- 2.7%), with higher results on questions related to more general EE answers (up to 81.0 +- 4.1%), and featuring promising multilingual abilities (4.4% accuracy loss due to translation).

Nicolò Oreste Pinciroli Vago, Francesca Forbicini, Piero Fraternali

Non-neural Machine Learning (ML) and Deep Learning (DL) models are often used to predict system failures in the context of industrial maintenance. However, only a few researches jointly assess the effect of varying the amount of past data used to make a prediction and the extension in the future of the forecast. This study evaluates the impact of the size of the reading window and of the prediction window on the performances of models trained to forecast failures in three data sets concerning the operation of (1) an industrial wrapping machine working in discrete sessions, (2) an industrial blood refrigerator working continuously, and (3) a nitrogen generator working continuously. The problem is formulated as a binary classification task that assigns the positive label to the prediction window based on the probability of a failure to occur in such an interval. Six algorithms (logistic regression, random forest, support vector machine, LSTM, ConvLSTM, and Transformers) are compared using multivariate telemetry time series. The results indicate that, in the considered scenarios, the dimension of the prediction windows plays a crucial role and highlight the effectiveness of DL approaches at classifying data with diverse time-dependent patterns preceding a failure and the effectiveness of ML approaches at classifying similar and repetitive patterns preceding a failure.

Francesca Forbicini, Nicolò Oreste Pinciroli Vago, Piero Fraternali

In both industrial and residential contexts, compressor-based machines, such as refrigerators, HVAC systems, heat pumps and chillers, are essential to fulfil production and consumers' needs. The diffusion of sensors and IoT connectivity supports the development of monitoring systems that can detect and predict faults, identify behavioural shifts and forecast the operational status of machines and their components. The focus of this paper is to survey the recent research on such tasks as FD, FP, Forecasting and CPD applied to multivariate time series characterizing the operations of compressor-based machines. These tasks play a critical role in improving the efficiency and longevity of machines by minimizing downtime and maintenance costs and improving the energy efficiency. Specifically, FD detects and diagnoses faults, FP predicts such occurrences, forecasting anticipates the future value of characteristic variables of machines and CPD identifies significant variations in the behaviour of the appliances, such as a change in the working regime. We identify and classify the approaches to the tasks mentioned above, compare the algorithms employed, highlight the gaps in the current status of the art and discuss the most promising future research directions in the field.

Nicolò Oreste Pinciroli Vago, Rocco Di Tella, Carolina Cuesta-Lázaro et al.

Astronomers have acquired vast repositories of multimodal data, including images, spectra, and time series, complemented by decades of literature that analyzes astrophysical sources. Still, these data sources are rarely systematically integrated. This work introduces a contrastive learning framework designed to align X-ray spectra with domain knowledge extracted from scientific literature, facilitating the development of shared multimodal representations. Establishing this connection is inherently complex, as scientific texts encompass a broader and more diverse physical context than spectra. We propose a contrastive pipeline that achieves a 20% Recall@1% when retrieving texts from spectra, proving that a meaningful alignment between these modalities is not only possible but capable of accelerating the interpretation of rare or poorly understood sources. Furthermore, the resulting shared latent space effectively encodes physically significant information. By fusing spectral and textual data, we improve the estimation of 20 physical variables by 16-18% over unimodal spectral baselines. Our results indicate that a Mixture of Experts (MoE) strategy, which leverages both unimodal and shared representations, yields superior performance. Finally, outlier analysis within the multimodal latent space identifies high-priority targets for follow-up investigation, including a candidate pulsating ULX (PULX) and a gravitational lens system. Importantly, this framework can be extended to other scientific domains where aligning observational data with existing literature is possible.

Alessandro Meroni, Nicolò Oreste Pinciroli Vago, Piero Fraternali

Estimating redshift is a central task in astrophysics, but its measurement is costly and time-consuming. In addition, current image-based methods are often validated on homogeneous datasets. The development and comparison of networks able generalize across different morphologies, ranging from galaxies to gravitationally-lensed transients, and observational conditions, remain an open challenge. This work proposes DeepRed, a deep learning pipeline that demonstrates how modern computer vision architectures, including ResNet, EfficientNet, Swin Transformer, and MLP-Mixer, can estimate redshifts from images of galaxies, gravitational lenses, and gravitationally-lensed supernovae. We compare these architectures and their ensemble to both neural networks (A1, A3, NetZ, and PhotoZ) and a feature-based method (HOG+SVR) on simulated (DeepGraviLens) and real (KiDS, SDSS) datasets. Our approach achieves state-of-the-art results on all datasets. On DeepGraviLens, DeepRed achieves a significant improvement in the Normalized Mean Absolute Deviation compared to the best baseline (PhotoZ): 55% on DES-deep (using EfficientNet), 51% on DES-wide (Ensemble), 52% on DESI-DOT (Ensemble), and 46% on LSST-wide (Ensemble). On real observations from the KiDS survey, the pipeline outperforms the best baseline (NetZ), improving NMAD by 16% on a general test set without high-probability lenses (Ensemble) and 27% on high-probability lenses (Ensemble). For non-lensed galaxies in the SDSS dataset, the MLP-Mixer architecture achieves a 5% improvement over the best baselines (A3 and NetZ). SHAP shows that the models correctly focus on the objects of interest with over 95% localization accuracy on high-quality images, validating the reliability of the predictions. These findings suggest that deep learning is a scalable, robust, and interpretable solution for redshift estimation in large-scale surveys.

Nicolò Oreste Pinciroli Vago, Juan Rafael Martínez-Galarza, Roberta Amato

The study of X-ray spectra is crucial to understanding the physical nature of astrophysical sources. Machine learning methods can extract compact and informative representations of data from large datasets. The Chandra Source Catalog (CSC) provides a rich archive of X-ray spectral data, which remains largely underexplored in this context. This work aims to develop a compact and physically meaningful representation of Chandra X-ray spectra using deep learning. To verify that the learned representation captures relevant information, we evaluate it through classification, regression, and interpretability analyses. We use a transformer-based autoencoder to compress X-ray spectra. The input spectra, drawn from the CSC, include only high-significance detections. Astrophysical source types and physical summary statistics are compiled from external catalogs. We evaluate the learned representation in terms of spectral reconstruction accuracy, clustering performance on 8 known astrophysical source classes, and correlation with physical quantities such as hardness ratios and hydrogen column density ($N_H$). The autoencoder accurately reconstructs spectra with 8 latent variables. Clustering in the latent space yields a balanced classification accuracy of $\sim$40% across the 8 source classes, increasing to $\sim$69% when restricted to AGNs and stellar-mass compact objects exclusively. Moreover, latent features correlate with non-linear combinations of spectral fluxes, suggesting that the compressed representation encodes physically relevant information. The proposed autoencoder-based pipeline is a powerful tool for the representation and interpretation of X-ray spectra, providing a compact latent space that supports both classification and the estimation of physical properties. This work demonstrates the potential of deep learning for spectral studies and uncovering new patterns in X-ray data.

Nicolò Oreste Pinciroli Vago, Roberta Amato, Matteo Imbrogno et al.

The discovery of fast and variable coherent signals in a handful of ultraluminous X-ray sources (ULXs) testifies to the presence of super-Eddington accreting neutron stars, and drastically changed the understanding of the ULX class. Our capability of discovering pulsations in ULXs is limited, among others, by poor statistics. However, catalogues and archives of high-energy missions contain information which can be used to identify new candidate pulsating ULXs (PULXs). The goal of this research is to single out candidate PULXs among those ULXs which have not shown pulsations due to an unfavourable combination of factors. We applied an AI approach to an updated database of ULXs detected by XMM-Newton. We first used an unsupervised clustering algorithm to sort out sources with similar characteristics into two clusters. Then, the sample of known PULX observations has been used to set the separation threshold between the two clusters and to identify the one containing the new candidate PULXs. We found that only a few criteria are needed to assign the membership of an observation to one of the two clusters. The cluster of new candidate PULXs counts 85 unique sources for 355 observations, with $\sim$85% of these new candidates having multiple observations. A preliminary timing analysis found no new pulsations for these candidates. This work presents a sample of new candidate PULXs observed by XMM-Newton, the properties of which are similar (in a multi-dimensional phase space) to those of the known PULXs, despite the absence of pulsations in their light curves. While this result is a clear example of the predictive power of AI-based methods, it also highlights the need for high-statistics observational data to reveal coherent signals from the sources in this sample and thus validate the robustness of the approach.

Nicolò Oreste Pinciroli Vago, Ibrahim A. Hameed, Michael Kachelriess

The presence of non-zero helicity in intergalactic magnetic fields is a smoking gun for their primordial origin since they have to be generated by processes that break CP invariance. As an experimental signature for the presence of helical magnetic fields, an estimator $Q$ based on the triple scalar product of the wave-vectors of photons generated in electromagnetic cascades from, e.g., TeV blazars, has been suggested previously. We propose to apply deep learning to helicity classification employing Convolutional Neural Networks and show that this method outperforms the $Q$ estimator.