Showing 1–50 of 85 results for author: Sanchez, D

Visual Grounding in Zero-Shot Vision-Language Control

Authors: J. de Curtò, Dayani Plasencia, Diego Sánchez, I. de Zarzà

Abstract: Vision-language models (VLMs) are increasingly used as zero-shot controllers, but successful trajectories do not necessarily show that decisions are grounded in visual input: simulator dynamics and conservative action priors can produce favourable scores without meaningful perception. We investigate this with an input-ablation battery: blind-image controls, repeated identical inputs, lane-axis ref… ▽ More Vision-language models (VLMs) are increasingly used as zero-shot controllers, but successful trajectories do not necessarily show that decisions are grounded in visual input: simulator dynamics and conservative action priors can produce favourable scores without meaningful perception. We investigate this with an input-ablation battery: blind-image controls, repeated identical inputs, lane-axis reflection, non-visual baselines, and pipeline-integrity checks. Across nine direct-action models, six structured local VLMs, and an exploratory VLM-MPC hierarchy, we analyse 32,874 scored calls over two embodiments and three simulators. The direct-control results are largely negative: a constant-SLOW policy outperforms a scripted geometric controller, several models are image-invariant or nearly constant, and models that recognize longitudinal hazards still fail to transform LEFT and RIGHT under reflection. No local VLM meets the joint longitudinal and lateral grounding criteria. However, an image-only deterministic positive control estimates the lead gap with 0.090 m MAE and exact mirror equivariance, confirming the stimuli carry sufficient visual information; the failures are modular, not universal. A post-hoc, leakage-controlled symmetry-consensus guardian selects two models from 16 calibration frames and freezes a 2-of-4 hazard vote across original and reflected views. On 272 held-out frames it reaches 0.954 balanced accuracy (episode-cluster bootstrap 95% CI [0.895,0.990]); nested leave-one-episode-out recovers the same pair and threshold in all 12 folds. Abstaining on ties raises committed balanced accuracy to 0.973 at 0.824 coverage. With deterministic perception retaining lateral authority, offline modular replay achieves 0.934 action agreement and exact mirror equivariance. These results support current VLMs as bounded, selective hazard assistants, not monolithic zero-shot controllers. △ Less

Submitted 6 August, 2026; originally announced August 2026.

Detecting Historical Turning Points in Italian Media: A Complex Systems Approach to a Diachronic News Corpus

Authors: Dario Zarcone, Salvatore Miccichè, David Sanchez

Abstract: The increasing availability of large-scale textual corpora has opened new possibilities for data-driven, quantitative approaches to historical analysis using Natural Language Processing (NLP). However, diachronic corpora with historical relevance from the pre-digital era remain scarce and often incomplete. We present a quantitative approach to historical analysis based on the reconstruction and ex… ▽ More The increasing availability of large-scale textual corpora has opened new possibilities for data-driven, quantitative approaches to historical analysis using Natural Language Processing (NLP). However, diachronic corpora with historical relevance from the pre-digital era remain scarce and often incomplete. We present a quantitative approach to historical analysis based on the reconstruction and exploration of a diachronic corpus of around 600,000 articles from the Italian newspaper "La Repubblica", covering all the articles published from the 1st of January 1985 to the 31st of December 2000 - a period of major political, social, and geopolitical change in Italy and globally. Using NLP techniques, we analyze the text at both lexical and semantic levels; we then apply tools from complex systems and statistical physics to trace shifts in media discourse over time. This allows us to detect key transition periods, such as the transition from the First Republic to the Second Republic in Italy, or major international conflicts like the Gulf War or the Kosovo War, without relying on prior labeling. The results show how combining computational linguistics with ideas from complex systems can offer new quantitative insight into historical changes, opening up new paths for studying the dynamics of media and society through large-scale textual data. △ Less

Submitted 12 June, 2026; originally announced June 2026.

Comments: 16 pages, 9 figures, 1 table

Harnessing AtomisticSkills for Agentic Atomistic Research

Authors: Bowen Deng, Bohan Li, Matthew Cox, Hoje Chun, Juno Nam, Artur Lyssenko, Sathya Edamadaka, Jurgis Ruza, Xiaochen Du, Nofit Segal, Jesus Diaz Sanchez, Mingrou Xie, Ty Perez, Yu Yao, Miguel Steiner, Sauradeep Majumdar, Charles B. Musgrave III, Anirban Chandra, Abhirup Patra, Detlef Hohl, Connor W. Coley, Ju Li, Rafael Gómez-Bombarelli

Abstract: Computational materials science and chemistry span vast knowledge domains and fractured software ecosystems. Although large language models (LLMs) have demonstrated research capabilities, scaling monolithic agents to manage the rigor and complexity of atomistic research remains a challenge. Here, we introduce AtomisticSkills, an open-source harness framework that empowers general-purpose AI coding… ▽ More Computational materials science and chemistry span vast knowledge domains and fractured software ecosystems. Although large language models (LLMs) have demonstrated research capabilities, scaling monolithic agents to manage the rigor and complexity of atomistic research remains a challenge. Here, we introduce AtomisticSkills, an open-source harness framework that empowers general-purpose AI coding agents to conduct atomistic research across materials science, chemistry, and drug discovery. By hierarchically decomposing scientific workflows into agent skills and tools, AtomisticSkills provides agents with modular, extensible, and plug-and-play research capabilities. The framework integrates more than 100 human-curated multidisciplinary skills, including database access, thermodynamics and kinetics modeling, and diverse simulation engines employing machine learning interatomic potentials (MLIPs) and density functional theory (DFT). We validate its functional coverage against scientific literature and demonstrate robust orchestration capabilities across diverse scientific campaigns: generative design of Li-ion solid-state electrolytes, high-throughput screening of metal-organic frameworks for CO2 capture, autonomous MLIP benchmarking and fine-tuning, multi-stage structure-based virtual screening for drug design, multimodal X-ray diffraction pattern analysis, and screening of Fe-oxide catalysts for oxygen evolution reaction. AtomisticSkills provides a critical agent infrastructure towards building fully autonomous AI scientists. △ Less

Submitted 18 May, 2026; originally announced May 2026.

From Knowledge to Action: Outcomes of the 2025 Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry

Authors: Aritra Roy, Kevin Shen, Andrew MacBride, Awwal Oladipupo, Mudassra Taskeen, Wojtek Treyde, Ruaa A. E. A. Abakar, Ahmad D. Abbas, Elsayed Abdelfatah, Abbas A. Abdullahi, Seham S. Abyah, Chahd Rahyl Adjmi, Fariha Agbere, Savyasanchi Aggarwal, Muhammad Ahmed, Tasnim Ahmed, Motasem Ajlouni, Mattias Akke, Hussein AlAdwan, Anwaar S. Alazani, Zahra A. Alharbi, Wajd A. Aljulyhi, Mohammed A. AlKubaish, Fatima A. Almahri, Sayed A. Almohri , et al. (328 additional authors not shown)

Abstract: Large language models (LLMs) are rapidly changing how researchers in materials science and chemistry discover, organize, and act on scientific knowledge. This paper analyzes a broad set of community-developed LLM applications in an effort to identify emerging patterns in how these systems can be used across the scientific research lifecycle. We organize the projects into two complementary categori… ▽ More Large language models (LLMs) are rapidly changing how researchers in materials science and chemistry discover, organize, and act on scientific knowledge. This paper analyzes a broad set of community-developed LLM applications in an effort to identify emerging patterns in how these systems can be used across the scientific research lifecycle. We organize the projects into two complementary categories: Knowledge Infrastructure, systems that structure, retrieve, synthesize, and validate scientific information; and Action Systems, systems that execute, coordinate, or automate scientific work across computational and experimental environments. The submissions reveal a shift from single-purpose LLM tools toward integrated, multi-agent workflows that combine retrieval, reasoning, tool use, and domain-specific validation. Prominent themes include retrieval-augmented generation as grounding infrastructure, persistent structured knowledge representations, multimodal and multilingual scientific inputs, and early progress toward laboratory-integrated closed-loop systems. Together, these results suggest that LLMs are evolving from general-purpose assistants into composable infrastructure for scientific reasoning and action. This work provides a community snapshot of that transition and a practical taxonomy for understanding emerging LLM-enabled workflows in materials science and chemistry. △ Less

Submitted 4 May, 2026; originally announced May 2026.

Comments: This paper reflects contributions from hundreds of researchers worldwide through an event, follow-on discussions, and project development exploring LLM applications in materials science and chemistry. While unconventional, it captures a timely, broad, and efficient community exploration of a rapidly evolving field and offers value to the arXiv community

Phonological distances for linguistic typology and the origin of Indo-European languages

Authors: Marius Mavridis, Juan De Gregorio, Raul Toral, David Sanchez

Abstract: We show that short-range phoneme dependencies encode large-scale patterns of linguistic relatedness, with direct implications for quantitative typology and evolutionary linguistics. Specifically, using an information-theoretic framework, we argue that phoneme sequences modeled as second-order Markov chains essentially capture the statistical correlations of a phonological system. This finding enab… ▽ More We show that short-range phoneme dependencies encode large-scale patterns of linguistic relatedness, with direct implications for quantitative typology and evolutionary linguistics. Specifically, using an information-theoretic framework, we argue that phoneme sequences modeled as second-order Markov chains essentially capture the statistical correlations of a phonological system. This finding enables us to quantify distances among 67 modern languages from a multilingual parallel corpus employing a distance metric that incorporates articulatory features of phonemes. The resulting phonological distance matrix recovers major language families and reveals signatures of contact-induced convergence. Remarkably, we obtain a clear correlation with geographic distance, allowing us to constrain a plausible homeland region for the Indo-European family, consistent with the Steppe hypothesis. △ Less

Submitted 13 April, 2026; originally announced April 2026.

Comments: 27 pages, 7 figures, 2 appendices

A Critical Review on the Effectiveness and Privacy Threats of Membership Inference Attacks

Authors: Najeeb Jebreel, David Sánchez, Josep Domingo-Ferrer

Abstract: Membership inference attacks (MIAs) aim to determine whether a data sample was included in a machine learning (ML) model's training set and have become the de facto standard for measuring privacy leakages in ML. We propose an evaluation framework that defines the conditions under which MIAs constitute a genuine privacy threat, and review representative MIAs against it. We find that, under the real… ▽ More Membership inference attacks (MIAs) aim to determine whether a data sample was included in a machine learning (ML) model's training set and have become the de facto standard for measuring privacy leakages in ML. We propose an evaluation framework that defines the conditions under which MIAs constitute a genuine privacy threat, and review representative MIAs against it. We find that, under the realistic conditions defined in our framework, MIAs represent weak privacy threats. Thus, relying on them as a privacy metric in ML can lead to an overestimation of risk and to unnecessary sacrifices in model utility as a consequence of employing too strong defenses. △ Less

Submitted 24 March, 2026; originally announced March 2026.

Comments: To appear in ESORICS 2026

Revisiting the LiRA Membership Inference Attack Under Realistic Assumptions

Authors: Najeeb Jebreel, Mona Khalil, David Sánchez, Josep Domingo-Ferrer

Abstract: Membership inference attacks (MIAs) have become the standard tool for evaluating privacy leakage in machine learning (ML). Among them, the Likelihood-Ratio Attack (LiRA) is widely regarded as the state of the art when sufficient shadow models are available. However, prior evaluations have often overstated the effectiveness of LiRA by attacking models overconfident on their training samples, calibr… ▽ More Membership inference attacks (MIAs) have become the standard tool for evaluating privacy leakage in machine learning (ML). Among them, the Likelihood-Ratio Attack (LiRA) is widely regarded as the state of the art when sufficient shadow models are available. However, prior evaluations have often overstated the effectiveness of LiRA by attacking models overconfident on their training samples, calibrating thresholds on target data, assuming balanced membership priors, and/or overlooking attack reproducibility. We re-evaluate LiRA under a realistic protocol that (i) trains models using anti-overfitting (AOF) and transfer learning (TL), when applicable, to reduce overconfidence as in production models; (ii) calibrates decision thresholds using shadow models and data rather than target data; (iii) measures positive predictive value (PPV, or precision) under shadow-based thresholds and skewed membership priors (pi <= 10%); and (iv) quantifies per-sample membership reproducibility across different seeds and training variations. We find that AOF significantly weakens LiRA, while TL further reduces attack effectiveness while improving model accuracy. Under shadow-based thresholds and skewed priors, LiRA's PPV often drops substantially, especially under AOF or AOF+TL. We also find that thresholded vulnerable sets at extremely low FPR show poor reproducibility across runs, while likelihood-ratio rankings are more stable. These results suggest that LiRA, and likely weaker MIAs, are less effective than previously suggested under realistic conditions, and that reliable privacy auditing requires evaluation protocols that reflect practical training practices, feasible attacker assumptions, and reproducibility considerations. Code is available at https://github.com/najeebjebreel/lira_analysis. △ Less

Submitted 8 March, 2026; originally announced March 2026.

Comments: Accepted to PoPETs 2026(3)

On the Practical Feasibility of Harvest-Now, Decrypt-Later Attacks

Authors: Javier Blanco-Romero, Florina Almenares Mendoza, Carlos García Rubio, Celeste Campo, Daniel Díaz Sánchez

Abstract: Harvest-now, decrypt-later (HN-DL) attacks threaten today's encrypted communications by archiving ciphertext until a quantum computer can break the underlying key exchange. This paper reframes HN-DL as an economic problem, quantifying adversary costs across Transport Layer Security (TLS) 1.2, TLS 1.3, QUIC, and Secure Shell (SSH) with an open-source testbed that reproduces the full attack sequence… ▽ More Harvest-now, decrypt-later (HN-DL) attacks threaten today's encrypted communications by archiving ciphertext until a quantum computer can break the underlying key exchange. This paper reframes HN-DL as an economic problem, quantifying adversary costs across Transport Layer Security (TLS) 1.2, TLS 1.3, QUIC, and Secure Shell (SSH) with an open-source testbed that reproduces the full attack sequence. Our model shows that retaining intercepted traffic is economically trivial, shifting the defensive question from whether an adversary can archive to how much decryption will cost. We evaluate protocol configuration strategies that act along two independent cost axes: storage overhead and quantum workload. Beyond the ongoing migration to post-quantum cryptography, these strategies provide defense in depth with current infrastructure. Encrypted Client Hello forces indiscriminate bulk collection, inflating the archive the adversary must retain, while aggressive rekeying and larger key exchange parameters multiply the quantum computations required to recover plaintext. Because storage inflation penalizes both sides while quantum cost inflation targets the adversary alone, rekeying and key size selection offer the strongest defensive levers. △ Less

Submitted 1 March, 2026; originally announced March 2026.

ODE-ViT: Plug & Play Attention Layer from the Generalization of the ViT as an Ordinary Differential Equation

Authors: Carlos Boned Riera, David Romero Sanchez, Oriol Ramos Terrades

Abstract: In recent years, increasingly large models have achieved outstanding performance across CV tasks. However, these models demand substantial computational resources and storage, and their growing complexity limits our understanding of how they make decisions. Most of these architectures rely on the attention mechanism within Transformer-based designs. Building upon the connection between residual ne… ▽ More In recent years, increasingly large models have achieved outstanding performance across CV tasks. However, these models demand substantial computational resources and storage, and their growing complexity limits our understanding of how they make decisions. Most of these architectures rely on the attention mechanism within Transformer-based designs. Building upon the connection between residual neural networks and ordinary differential equations (ODEs), we introduce ODE-ViT, a Vision Transformer reformulated as an ODE system that satisfies the conditions for well-posed and stable dynamics. Experiments on CIFAR-10 and CIFAR-100 demonstrate that ODE-ViT achieves stable, interpretable, and competitive performance with up to one order of magnitude fewer parameters, surpassing prior ODE-based Transformer approaches in classification tasks. We further propose a plug-and-play teacher-student framework in which a discrete ViT guides the continuous trajectory of ODE-ViT by treating the intermediate representations of the teacher as solutions of the ODE. This strategy improves performance by more than 10% compared to training a free ODE-ViT from scratch. △ Less

Submitted 20 November, 2025; originally announced November 2025.

FAIRPLAI: A Human-in-the-Loop Approach to Fair and Private Machine Learning

Authors: David Sanchez Jr., Holly Lopez, Michelle Buraczyk, Anantaa Kotal

Abstract: As machine learning systems move from theory to practice, they are increasingly tasked with decisions that affect healthcare access, financial opportunities, hiring, and public services. In these contexts, accuracy is only one piece of the puzzle - models must also be fair to different groups, protect individual privacy, and remain accountable to stakeholders. Achieving all three is difficult: dif… ▽ More As machine learning systems move from theory to practice, they are increasingly tasked with decisions that affect healthcare access, financial opportunities, hiring, and public services. In these contexts, accuracy is only one piece of the puzzle - models must also be fair to different groups, protect individual privacy, and remain accountable to stakeholders. Achieving all three is difficult: differential privacy can unintentionally worsen disparities, fairness interventions often rely on sensitive data that privacy restricts, and automated pipelines ignore that fairness is ultimately a human and contextual judgment. We introduce FAIRPLAI (Fair and Private Learning with Active Human Influence), a practical framework that integrates human oversight into the design and deployment of machine learning systems. FAIRPLAI works in three ways: (1) it constructs privacy-fairness frontiers that make trade-offs between accuracy, privacy guarantees, and group outcomes transparent; (2) it enables interactive stakeholder input, allowing decision-makers to select fairness criteria and operating points that reflect their domain needs; and (3) it embeds a differentially private auditing loop, giving humans the ability to review explanations and edge cases without compromising individual data security. Applied to benchmark datasets, FAIRPLAI consistently preserves strong privacy protections while reducing fairness disparities relative to automated baselines. More importantly, it provides a straightforward, interpretable process for practitioners to manage competing demands of accuracy, privacy, and fairness in socially impactful applications. By embedding human judgment where it matters most, FAIRPLAI offers a pathway to machine learning systems that are effective, responsible, and trustworthy in practice. GitHub: https://github.com/Li1Davey/Fairplai △ Less

Submitted 11 November, 2025; originally announced November 2025.

How to Get Actual Privacy and Utility from Privacy Models: the k-Anonymity and Differential Privacy Families

Authors: Josep Domingo-Ferrer, David Sánchez

Abstract: Privacy models were introduced in privacy-preserving data publishing and statistical disclosure control with the promise to end the need for costly empirical assessment of disclosure risk. We examine how well this promise is kept by the main privacy models. We find they may fail to provide adequate protection guarantees because of problems in their definition or incur unacceptable trade-offs betwe… ▽ More Privacy models were introduced in privacy-preserving data publishing and statistical disclosure control with the promise to end the need for costly empirical assessment of disclosure risk. We examine how well this promise is kept by the main privacy models. We find they may fail to provide adequate protection guarantees because of problems in their definition or incur unacceptable trade-offs between privacy protection and utility preservation. Specifically, k-anonymity may not entirely exclude disclosure if enforced with deterministic mechanisms or without constraints on the confidential values. On the other hand, differential privacy (DP) incurs unacceptable utility loss for small budgets and its privacy guarantee becomes meaningless for large budgets. In the latter case, an ex post empirical assessment of disclosure risk becomes necessary, undermining the main appeal of privacy models. Whereas the utility preservation of DP can only be improved by relaxing its privacy guarantees, we argue that a semantic reformulation of k-anonymity can offer more robust privacy without losing utility with respect to traditional syntactic k-anonymity. △ Less

Submitted 17 October, 2025; v1 submitted 13 October, 2025; originally announced October 2025.

Comments: 13 pages

MSC Class: 68 ACM Class: K.4.1

Information-theoretic analysis of temporal dependence in discrete stochastic processes: Application to precipitation predictability

Authors: Juan De Gregorio, David Sánchez, Raúl Toral

Abstract: Understanding the temporal dependence of precipitation is key to improving weather predictability and developing efficient stochastic rainfall models. We introduce an information-theoretic approach to quantify memory effects in discrete stochastic processes and apply it to daily precipitation records across the contiguous United States. The method is based on the predictability gain, a quantity de… ▽ More Understanding the temporal dependence of precipitation is key to improving weather predictability and developing efficient stochastic rainfall models. We introduce an information-theoretic approach to quantify memory effects in discrete stochastic processes and apply it to daily precipitation records across the contiguous United States. The method is based on the predictability gain, a quantity derived from block entropy that measures the additional information provided by higher-order temporal dependencies. This statistic, combined with a bootstrap-based hypothesis testing and Fisher's method, enables a robust memory estimator from finite data. Tests with generated sequences show that this estimator outperforms other model-selection criteria such as Akaike Information Criterion and Bayesian Information Criterion. Applied to precipitation data, the analysis reveals that daily rainfall occurrence is well described by low-order Markov chains, exhibiting regional and seasonal variations, with stronger correlations in winter along the West Coast and in summer in the Southeast, consistent with known climatological patterns. Overall, our findings establish a framework for building parsimonious stochastic descriptions, useful when addressing spatial heterogeneity in the memory structure of precipitation dynamics, and support further advances in real-time, data-driven forecasting schemes. △ Less

Submitted 13 March, 2026; v1 submitted 13 October, 2025; originally announced October 2025.

Journal ref: Chaos 36, 033124 (1-14) (2026)

Preserving Russek's "Summermood" Using Reality Check and a DeltaLab DL-4 Approximation

Authors: Jeremy Hyrkas, Pablo Dodero Carrillo, Teresa Díaz de Cossio Sánchez

Abstract: As a contribution towards ongoing efforts to maintain electroacoustic compositions for live performance, we present a collection of Pure Data patches to preserve and perform Antonio Russek's piece "Summermood" for bass flute and live electronics. The piece, originally written for the DeltaLab DL-4 delay rack unit, contains score markings specific to the DL-4. Here, we approximate the sound and uni… ▽ More As a contribution towards ongoing efforts to maintain electroacoustic compositions for live performance, we present a collection of Pure Data patches to preserve and perform Antonio Russek's piece "Summermood" for bass flute and live electronics. The piece, originally written for the DeltaLab DL-4 delay rack unit, contains score markings specific to the DL-4. Here, we approximate the sound and unique functionality of the DL-4 in Pure Data, then refine our implementation to better match the unit on which the piece was performed by comparing settings from the score to two official recordings of the piece. The DL-4 emulation is integrated into a patch for live performance based on the Null Piece, and regression tested using the Reality Check framework for Pure Data. Using this library of patches, Summermood can be brought back into live rotation without the use of the now discontinued DL-4. The patches will be continuously tested to ensure that the piece is playable across computer environments and as the Pure Data programming language is updated. △ Less

Submitted 25 September, 2025; originally announced September 2025.

Comments: 6 pages, 10 figures, Pure Data Max Conference 2025

Journal ref: Proceedings and Programs of PdMaxCon25~ (2025) 55-60

WATCHED: A Web AI Agent Tool for Combating Hate Speech by Expanding Data

Authors: Paloma Piot, Diego Sánchez, Javier Parapar

Abstract: Online harms are a growing problem in digital spaces, putting user safety at risk and reducing trust in social media platforms. One of the most persistent forms of harm is hate speech. To address this, we need tools that combine the speed and scale of automated systems with the judgment and insight of human moderators. These tools should not only find harmful content but also explain their decisio… ▽ More Online harms are a growing problem in digital spaces, putting user safety at risk and reducing trust in social media platforms. One of the most persistent forms of harm is hate speech. To address this, we need tools that combine the speed and scale of automated systems with the judgment and insight of human moderators. These tools should not only find harmful content but also explain their decisions clearly, helping to build trust and understanding. In this paper, we present WATCHED, a chatbot designed to support content moderators in tackling hate speech. The chatbot is built as an Artificial Intelligence Agent system that uses Large Language Models along with several specialised tools. It compares new posts with real examples of hate speech and neutral content, uses a BERT-based classifier to help flag harmful messages, looks up slang and informal language using sources like Urban Dictionary, generates chain-of-thought reasoning, and checks platform guidelines to explain and support its decisions. This combination allows the chatbot not only to detect hate speech but to explain why content is considered harmful, grounded in both precedent and policy. Experimental results show that our proposed method surpasses existing state-of-the-art methods, reaching a macro F1 score of 0.91. Designed for moderators, safety teams, and researchers, the tool helps reduce online harms by supporting collaboration between AI and human oversight. △ Less

Submitted 1 September, 2025; originally announced September 2025.

Efficient Unlearning with Privacy Guarantees

Authors: Josep Domingo-Ferrer, Najeeb Jebreel, David Sánchez

Abstract: Privacy protection laws, such as the GDPR, grant individuals the right to request the forgetting of their personal data not only from databases but also from machine learning (ML) models trained on them. Machine unlearning has emerged as a practical means to facilitate model forgetting of data instances seen during training. Although some existing machine unlearning methods guarantee exact forgett… ▽ More Privacy protection laws, such as the GDPR, grant individuals the right to request the forgetting of their personal data not only from databases but also from machine learning (ML) models trained on them. Machine unlearning has emerged as a practical means to facilitate model forgetting of data instances seen during training. Although some existing machine unlearning methods guarantee exact forgetting, they are typically costly in computational terms. On the other hand, more affordable methods do not offer forgetting guarantees and are applicable only to specific ML models. In this paper, we present \emph{efficient unlearning with privacy guarantees} (EUPG), a novel machine unlearning framework that offers formal privacy guarantees to individuals whose data are being unlearned. EUPG involves pre-training ML models on data protected using privacy models, and it enables {\em efficient unlearning with the privacy guarantees offered by the privacy models in use}. Through empirical evaluation on four heterogeneous data sets protected with $k$-anonymity and $ε$-differential privacy as privacy models, our approach demonstrates utility and forgetting effectiveness comparable to those of exact unlearning methods, while significantly reducing computational and storage costs. Our code is available at https://github.com/najeebjebreel/EUPG. △ Less

Submitted 26 June, 2026; v1 submitted 7 July, 2025; originally announced July 2025.

Comments: 34 pages, 9 tables, 2 figures

LiDAR-Guided Monocular 3D Object Detection for Long-Range Railway Monitoring

Authors: Raul David Dominguez Sanchez, Xavier Diaz Ortiz, Xingcheng Zhou, Max Peter Ronecker, Michael Karner, Daniel Watzenig, Alois Knoll

Abstract: Railway systems, particularly in Germany, require high levels of automation to address legacy infrastructure challenges and increase train traffic safely. A key component of automation is robust long-range perception, essential for early hazard detection, such as obstacles at level crossings or pedestrians on tracks. Unlike automotive systems with braking distances of ~70 meters, trains require pe… ▽ More Railway systems, particularly in Germany, require high levels of automation to address legacy infrastructure challenges and increase train traffic safely. A key component of automation is robust long-range perception, essential for early hazard detection, such as obstacles at level crossings or pedestrians on tracks. Unlike automotive systems with braking distances of ~70 meters, trains require perception ranges exceeding 1 km. This paper presents an deep-learning-based approach for long-range 3D object detection tailored for autonomous trains. The method relies solely on monocular images, inspired by the Faraway-Frustum approach, and incorporates LiDAR data during training to improve depth estimation. The proposed pipeline consists of four key modules: (1) a modified YOLOv9 for 2.5D object detection, (2) a depth estimation network, and (3-4) dedicated short- and long-range 3D detection heads. Evaluations on the OSDaR23 dataset demonstrate the effectiveness of the approach in detecting objects up to 250 meters. Results highlight its potential for railway automation and outline areas for future improvement. △ Less

Submitted 25 April, 2025; originally announced April 2025.

Comments: Accepted for the Data-Driven Learning for Intelligent Vehicle Applications Workshop at the 36th IEEE Intelligent Vehicles Symposium (IV) 2025

DP2Unlearning: An Efficient and Guaranteed Unlearning Framework for LLMs

Authors: Tamim Al Mahmud, Najeeb Jebreel, Josep Domingo-Ferrer, David Sanchez

Abstract: Large language models (LLMs) have recently revolutionized language processing tasks but have also brought ethical and legal issues. LLMs have a tendency to memorize potentially private or copyrighted information present in the training data, which might then be delivered to end users at inference time. When this happens, a naive solution is to retrain the model from scratch after excluding the und… ▽ More Large language models (LLMs) have recently revolutionized language processing tasks but have also brought ethical and legal issues. LLMs have a tendency to memorize potentially private or copyrighted information present in the training data, which might then be delivered to end users at inference time. When this happens, a naive solution is to retrain the model from scratch after excluding the undesired data. Although this guarantees that the target data have been forgotten, it is also prohibitively expensive for LLMs. Approximate unlearning offers a more efficient alternative, as it consists of ex post modifications of the trained model itself to prevent undesirable results, but it lacks forgetting guarantees because it relies solely on empirical evidence. In this work, we present DP2Unlearning, a novel LLM unlearning framework that offers formal forgetting guarantees at a significantly lower cost than retraining from scratch on the data to be retained. DP2Unlearning involves training LLMs on textual data protected using ε-differential privacy (DP), which later enables efficient unlearning with the guarantees against disclosure associated with the chosen ε. Our experiments demonstrate that DP2Unlearning achieves similar model performance post-unlearning, compared to an LLM retraining from scratch on retained data -- the gold standard exact unlearning -- but at approximately half the unlearning cost. In addition, with a reasonable computational cost, it outperforms approximate unlearning methods at both preserving the utility of the model post-unlearning and effectively forgetting the targeted information. △ Less

Submitted 18 July, 2025; v1 submitted 18 April, 2025; originally announced April 2025.

Comments: This is the updated version of the preprint, revised following acceptance for publication in Elsevier Neural Networks Journal. The paper is now published (18 July 2025) with DOI: https://doi.org/10.1016/j.neunet.2025.107879

Journal ref: Neural Networks, 2025, Article 107879

RigoChat 2: an adapted language model to Spanish using a bounded dataset and reduced hardware

Authors: Gonzalo Santamaría Gómez, Guillem García Subies, Pablo Gutiérrez Ruiz, Mario González Valero, Natàlia Fuertes, Helena Montoro Zamorano, Carmen Muñoz Sanz, Leire Rosado Plaza, Nuria Aldama García, David Betancur Sánchez, Kateryna Sushkova, Marta Guerrero Nieto, Álvaro Barbero Jiménez

Abstract: Large Language Models (LLMs) have become a key element of modern artificial intelligence, demonstrating the ability to address a wide range of language processing tasks at unprecedented levels of accuracy without the need of collecting problem-specific data. However, these versatile models face a significant challenge: both their training and inference processes require substantial computational r… ▽ More Large Language Models (LLMs) have become a key element of modern artificial intelligence, demonstrating the ability to address a wide range of language processing tasks at unprecedented levels of accuracy without the need of collecting problem-specific data. However, these versatile models face a significant challenge: both their training and inference processes require substantial computational resources, time, and memory. Consequently, optimizing this kind of models to minimize these requirements is crucial. In this article, we demonstrate that, with minimal resources and in a remarkably short time, it is possible to enhance a state-of-the-art model, specifically for a given language task, without compromising its overall capabilities using a relatively small pretrained LLM as a basis. Specifically, we present our use case, RigoChat 2, illustrating how LLMs can be adapted to achieve superior results in Spanish-language tasks. △ Less

Submitted 11 March, 2025; originally announced March 2025.

MEL: Legal Spanish Language Model

Authors: David Betancur Sánchez, Nuria Aldama García, Álvaro Barbero Jiménez, Marta Guerrero Nieto, Patricia Marsà Morales, Nicolás Serrano Salas, Carlos García Hernán, Pablo Haya Coll, Elena Montiel Ponsoda, Pablo Calleja Ibáñez

Abstract: Legal texts, characterized by complex and specialized terminology, present a significant challenge for Language Models. Adding an underrepresented language, such as Spanish, to the mix makes it even more challenging. While pre-trained models like XLM-RoBERTa have shown capabilities in handling multilingual corpora, their performance on domain specific documents remains underexplored. This paper pr… ▽ More Legal texts, characterized by complex and specialized terminology, present a significant challenge for Language Models. Adding an underrepresented language, such as Spanish, to the mix makes it even more challenging. While pre-trained models like XLM-RoBERTa have shown capabilities in handling multilingual corpora, their performance on domain specific documents remains underexplored. This paper presents the development and evaluation of MEL, a legal language model based on XLM-RoBERTa-large, fine-tuned on legal documents such as BOE (Boletín Oficial del Estado, the Spanish oficial report of laws) and congress texts. We detail the data collection, processing, training, and evaluation processes. Evaluation benchmarks show a significant improvement over baseline models in understanding the legal Spanish language. We also present case studies demonstrating the model's application to new legal texts, highlighting its potential to perform top results over different NLP tasks. △ Less

Submitted 27 January, 2025; originally announced January 2025.

Comments: 8 pages, 6 figures, 3 tables

3CEL: A corpus of legal Spanish contract clauses

Authors: Nuria Aldama García, Patricia Marsà Morales, David Betancur Sánchez, Álvaro Barbero Jiménez, Marta Guerrero Nieto, Pablo Haya Coll, Patricia Martín Chozas, Elena Montiel Ponsoda

Abstract: Legal corpora for Natural Language Processing (NLP) are valuable and scarce resources in languages like Spanish due to two main reasons: data accessibility and legal expert knowledge availability. INESData 2024 is a European Union funded project lead by the Universidad Politécnica de Madrid (UPM) and developed by Instituto de Ingeniería del Conocimiento (IIC) to create a series of state-of-the-art… ▽ More Legal corpora for Natural Language Processing (NLP) are valuable and scarce resources in languages like Spanish due to two main reasons: data accessibility and legal expert knowledge availability. INESData 2024 is a European Union funded project lead by the Universidad Politécnica de Madrid (UPM) and developed by Instituto de Ingeniería del Conocimiento (IIC) to create a series of state-of-the-art NLP resources applied to the legal/administrative domain in Spanish. The goal of this paper is to present the Corpus of Legal Spanish Contract Clauses (3CEL), which is a contract information extraction corpus developed within the framework of INESData 2024. 3CEL contains 373 manually annotated tenders using 19 defined categories (4 782 total tags) that identify key information for contract understanding and reviewing. △ Less

Submitted 27 January, 2025; originally announced January 2025.

Comments: 12 pages, 13 figures, 6 tables

Motives meet SymPy: studying $λ$-ring expressions in Python

Authors: Daniel Sanchez, David Alfaya, Jaime Pizarroso

Abstract: We present a new Python package called "motives", a symbolic manipulation package based on SymPy capable of handling and simplifying motivic expressions in the Grothendieck ring of Chow motives and other types of $λ$-rings. The package is able to manipulate and compare arbitrary expressions in $λ$-rings and, in particular, it contains explicit tools for manipulating motives of several types of com… ▽ More We present a new Python package called "motives", a symbolic manipulation package based on SymPy capable of handling and simplifying motivic expressions in the Grothendieck ring of Chow motives and other types of $λ$-rings. The package is able to manipulate and compare arbitrary expressions in $λ$-rings and, in particular, it contains explicit tools for manipulating motives of several types of commonly used moduli schemes and moduli stacks of decorated bundles on curves. We have applied this new tool to advance in the verification of Mozgovoy's conjectural formula for the motive of the moduli space of twisted Higgs bundles, proving that it holds in rank 2 and 3 for any curve of genus up to 18 and any twisting bundle of small degree. △ Less

Submitted 31 December, 2024; originally announced January 2025.

Comments: 19 pages, 2 figures. The code of the library described in the paper is publicly hosted at https://github.com/CIAMOD/motives

MSC Class: 13D15 (Primary) 68W30; 19E08; 14C35; 14D20; 14H60 (Secondary) ACM Class: I.1.1

Truthful Text Sanitization Guided by Inference Attacks

Authors: Ildikó Pilán, Benet Manzanares-Salor, David Sánchez, Pierre Lison

Abstract: Text sanitization aims to rewrite parts of a document to prevent disclosure of personal information. The central challenge of text sanitization is to strike a balance between privacy protection (avoiding the leakage of personal information) and utility preservation (retaining as much as possible of the document's original content). To this end, we introduce a novel text sanitization method based o… ▽ More Text sanitization aims to rewrite parts of a document to prevent disclosure of personal information. The central challenge of text sanitization is to strike a balance between privacy protection (avoiding the leakage of personal information) and utility preservation (retaining as much as possible of the document's original content). To this end, we introduce a novel text sanitization method based on generalizations, that is, broader but still informative terms that subsume the semantic content of the original text spans. The approach relies on the use of instruction-tuned large language models (LLMs) and is divided into two stages. Given a document including text spans expressing personally identifiable information (PII), the LLM is first applied to obtain truth-preserving replacement candidates for each text span and rank those according to their abstraction level. Those candidates are then evaluated for their ability to protect privacy by conducting inference attacks with the LLM. Finally, the system selects the most informative replacement candidate shown to be resistant to those attacks. This two-stage process produces replacements that effectively balance privacy and utility. We also present novel metrics to evaluate these two aspects without needing to manually annotate documents. Results on the Text Anonymization Benchmark show that the proposed approach, implemented with Mistral 7B Instruct, leads to enhanced utility, with only a marginal (< 1 p.p.) increase in re-identification risk compared to fully suppressing the original spans. Furthermore, our approach is shown to be more truth-preserving than existing methods such as Microsoft Presidio's synthetic replacements. △ Less

Submitted 31 August, 2025; v1 submitted 17 December, 2024; originally announced December 2024.

Enhanced 5G/B5G Network Planning/Optimization deploying RIS in Urban/Outdoor Scenarios

Authors: Valdemar Farré, Juan C. Estrada-Jiménez, José D. Vega Sánchez, Juan A. Vasquez-Peralvo, Symeon Chatzinotas

Abstract: In recent years, the fifth-generation (5G) mobile network has been developed worldwide to remarkably improve network performance and spectral efficiency. Very recently, reconfigurable intelligent surfaces (RISs) technology has emerged as an innovative solution for controlling the propagation medium of the forthcoming sixth-generation (6G) networks. Specifically, RIS takes advantage of the reflecte… ▽ More In recent years, the fifth-generation (5G) mobile network has been developed worldwide to remarkably improve network performance and spectral efficiency. Very recently, reconfigurable intelligent surfaces (RISs) technology has emerged as an innovative solution for controlling the propagation medium of the forthcoming sixth-generation (6G) networks. Specifically, RIS takes advantage of the reflected rays on the propagation environment to redirect them to a desired target, improving wireless coverage. To further improve RIS performance, an interesting technique called synchronized transmission with advanced reconfigurable surfaces (STARS) has appeared to allow simultaneous transmission and reflection of intelligent omni-surfaces. With that in mind, this paper introduces an enhanced strategy for the network planning of 5G and beyond (B5G) mobile networks in dense urban scenarios focused on the city of Quito-Ecuador. The efficacy of RIS and its cutting-edge STARS concept is emphasized, providing useful insights into coverage, quality, and throughput results. In particular, this work considers the 3.5/28 GHz frequency bands, optimizing the radio network and anticipating their applicability in B5G networks. Finally, simulation results are also shown, which allow the identification of the benefits of STAR RIS in terms of coverage, signal quality, and data performance. △ Less

Submitted 6 December, 2024; originally announced December 2024.

Comments: 6 pages, 12 figures, 2024 IEEE 29th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks (CAMAD), presented in Session 11: Federation of 6G Infrastructures and Experimentation Facilities in a Network of Networks, October 22th - 2024

Entropy and type-token ratio in gigaword corpora

Authors: Pablo Rosillo-Rodes, Maxi San Miguel, David Sanchez

Abstract: There are different ways of measuring diversity in complex systems. In particular, in language, lexical diversity is characterized in terms of the type-token ratio and the word entropy. We here investigate both diversity metrics in six massive linguistic datasets in English, Spanish, and Turkish, consisting of books, news articles, and tweets. These gigaword corpora correspond to languages with di… ▽ More There are different ways of measuring diversity in complex systems. In particular, in language, lexical diversity is characterized in terms of the type-token ratio and the word entropy. We here investigate both diversity metrics in six massive linguistic datasets in English, Spanish, and Turkish, consisting of books, news articles, and tweets. These gigaword corpora correspond to languages with distinct morphological features and differ in registers and genres, thus constituting a varied testbed for a quantitative approach to lexical diversity. We unveil an empirical functional relation between entropy and type-token ratio of texts of a given corpus and language, which is a consequence of the statistical laws observed in natural language. Further, in the limit of large text lengths we find an analytical expression for this relation relying on both Zipf and Heaps laws that agrees with our empirical findings. △ Less

Submitted 24 June, 2025; v1 submitted 15 November, 2024; originally announced November 2024.

Comments: 15 pages, 10 figures, 8 tables

Journal ref: Phys. Rev. Research 7, 033054 (2025)

Computational lexical analysis of Flamenco genres

Authors: Pablo Rosillo-Rodes, Maxi San Miguel, David Sanchez

Abstract: Flamenco, recognized by UNESCO as part of the Intangible Cultural Heritage of Humanity, is a profound expression of cultural identity rooted in Andalusia, Spain. However, there is a lack of quantitative studies that help identify characteristic patterns in this long-lived music tradition. In this work, we present a computational analysis of Flamenco lyrics, employing natural language processing an… ▽ More Flamenco, recognized by UNESCO as part of the Intangible Cultural Heritage of Humanity, is a profound expression of cultural identity rooted in Andalusia, Spain. However, there is a lack of quantitative studies that help identify characteristic patterns in this long-lived music tradition. In this work, we present a computational analysis of Flamenco lyrics, employing natural language processing and machine learning to categorize over 2000 lyrics into their respective Flamenco genres, termed as $\textit{palos}$. Using a Multinomial Naive Bayes classifier, we find that lexical variation across styles enables to accurately identify distinct $\textit{palos}$. More importantly, from an automatic method of word usage, we obtain the semantic fields that characterize each style. Further, applying a metric that quantifies the inter-genre distance we perform a network analysis that sheds light on the relationship between Flamenco styles. Remarkably, our results suggest historical connections and $\textit{palo}$ evolutions. Overall, our work illuminates the intricate relationships and cultural significance embedded within Flamenco lyrics, complementing previous qualitative discussions with quantitative analyses and sparking new discussions on the origin and development of traditional music genres. △ Less

Submitted 13 March, 2026; v1 submitted 9 May, 2024; originally announced May 2024.

Comments: 25 pages, 20 figures

Journal ref: ACM J. Comput. Cult. Herit. 18, 59 (2025)

Digital Forgetting in Large Language Models: A Survey of Unlearning Methods

Authors: Alberto Blanco-Justicia, Najeeb Jebreel, Benet Manzanares, David Sánchez, Josep Domingo-Ferrer, Guillem Collell, Kuan Eeik Tan

Abstract: The objective of digital forgetting is, given a model with undesirable knowledge or behavior, obtain a new model where the detected issues are no longer present. The motivations for forgetting include privacy protection, copyright protection, elimination of biases and discrimination, and prevention of harmful content generation. Effective digital forgetting has to be effective (meaning how well th… ▽ More The objective of digital forgetting is, given a model with undesirable knowledge or behavior, obtain a new model where the detected issues are no longer present. The motivations for forgetting include privacy protection, copyright protection, elimination of biases and discrimination, and prevention of harmful content generation. Effective digital forgetting has to be effective (meaning how well the new model has forgotten the undesired knowledge/behavior), retain the performance of the original model on the desirable tasks, and be scalable (in particular forgetting has to be more efficient than retraining from scratch on just the tasks/data to be retained). This survey focuses on forgetting in large language models (LLMs). We first provide background on LLMs, including their components, the types of LLMs, and their usual training pipeline. Second, we describe the motivations, types, and desired properties of digital forgetting. Third, we introduce the approaches to digital forgetting in LLMs, among which unlearning methodologies stand out as the state of the art. Fourth, we provide a detailed taxonomy of machine unlearning methods for LLMs, and we survey and compare current approaches. Fifth, we detail datasets, models and metrics used for the evaluation of forgetting, retaining and runtime. Sixth, we discuss challenges in the area. Finally, we provide some concluding remarks. △ Less

Submitted 2 April, 2024; originally announced April 2024.

Comments: 70 pages

MSC Class: 68 ACM Class: K.4.1; I.2.6; I.2.7

Journal ref: Artificial Intelligence Review, vol. 58, art. no. 90, 2025

Exploring language relations through syntactic distances and geographic proximity

Authors: Juan De Gregorio, Raúl Toral, David Sánchez

Abstract: Languages are grouped into families that share common linguistic traits. While this approach has been successful in understanding genetic relations between diverse languages, more analyses are needed to accurately quantify their relatedness, especially in less studied linguistic levels such as syntax. Here, we explore linguistic distances using series of parts of speech (POS) extracted from the Un… ▽ More Languages are grouped into families that share common linguistic traits. While this approach has been successful in understanding genetic relations between diverse languages, more analyses are needed to accurately quantify their relatedness, especially in less studied linguistic levels such as syntax. Here, we explore linguistic distances using series of parts of speech (POS) extracted from the Universal Dependencies dataset. Within an information-theoretic framework, we show that employing POS trigrams maximizes the possibility of capturing syntactic variations while being at the same time compatible with the amount of available data. Linguistic connections are then established by assessing pairwise distances based on the POS distributions. Intriguingly, our analysis reveals definite clusters that correspond to well known language families and groups, with exceptions explained by distinct morphological typologies. Furthermore, we obtain a significant correlation between language similarity and geographic distance, which underscores the influence of spatial proximity on language kinships. △ Less

Submitted 3 October, 2024; v1 submitted 27 March, 2024; originally announced March 2024.

Comments: 39 pages

Journal ref: EPJ Data Science 13, 61 (2024)

The socialisation of the adolescent who carries out team sports: a transversal study of centrality with a social network analysis

Authors: Pilar Marqués-Sánchez, José Alberto Benítez-Andrades, María Dolores Calvo Sánchez, Natalia Arias

Abstract: Objectives: This study analyzed adolescent physical activity, its link to overweight, and the social network structure in group sports participants, focusing on centrality measures. Setting: Conducted in 11 classrooms across 5 schools in Ponferrada, Spain. Participants: Included 235 adolescents (49.4% female), categorized as normal weight or overweight. Methods: The Physical Activity Questio… ▽ More Objectives: This study analyzed adolescent physical activity, its link to overweight, and the social network structure in group sports participants, focusing on centrality measures. Setting: Conducted in 11 classrooms across 5 schools in Ponferrada, Spain. Participants: Included 235 adolescents (49.4% female), categorized as normal weight or overweight. Methods: The Physical Activity Questionnaire for Adolescents (PAQ-A) assessed physical activity levels. Social network analysis evaluated centrality in varying contact degrees. Results: 30.2% were overweight. Males scored higher in PAQ-A and were more likely to engage in group sports. No significant correlation was found between physical activity and weight in the total sample. However, overweight females reported higher exercise levels. Centrality analysis showed gender differences; women in group sports had lower centrality, whereas men had higher. Conclusions: The study highlights the importance of gender and social network centrality in designing future strategies, considering peer interaction intensity △ Less

Submitted 14 February, 2024; originally announced February 2024.

Journal ref: BMJ Open, 2021, Volume 11, ID e042773

Conciliating Privacy and Utility in Data Releases via Individual Differential Privacy and Microaggregation

Authors: Jordi Soria-Comas, David Sánchez, Josep Domingo-Ferrer, Sergio Martínez, Luis Del Vasto-Terrientes

Abstract: $ε$-Differential privacy (DP) is a well-known privacy model that offers strong privacy guarantees. However, when applied to data releases, DP significantly deteriorates the analytical utility of the protected outcomes. To keep data utility at reasonable levels, practical applications of DP to data releases have used weak privacy parameters (large $ε… ▽ More $ε$-Differential privacy (DP) is a well-known privacy model that offers strong privacy guarantees. However, when applied to data releases, DP significantly deteriorates the analytical utility of the protected outcomes. To keep data utility at reasonable levels, practical applications of DP to data releases have used weak privacy parameters (large $ε$), which dilute the privacy guarantees of DP. In this work, we tackle this issue by using an alternative formulation of the DP privacy guarantees, named $ε$-individual differential privacy (iDP), which causes less data distortion while providing the same protection as DP to subjects. We enforce iDP in data releases by relying on attribute masking plus a pre-processing step based on data microaggregation. The goal of this step is to reduce the sensitivity to record changes, which determines the amount of noise required to enforce iDP (and DP). Specifically, we propose data microaggregation strategies designed for iDP whose sensitivities are significantly lower than those used in DP. As a result, we obtain iDP-protected data with significantly better utility than with DP. We report on experiments that show how our approach can provide strong privacy (small $ε$) while yielding protected data that do not significantly degrade the accuracy of secondary data analysis. △ Less

Submitted 21 December, 2023; originally announced December 2023.

Comments: 17 pages, 6 figures

Multi-Task Faces (MTF) Data Set: A Legally and Ethically Compliant Collection of Face Images for Various Classification Tasks

Authors: Rami Haffar, David Sánchez, Josep Domingo-Ferrer

Abstract: Human facial data offers valuable potential for tackling classification problems, including face recognition, age estimation, gender identification, emotion analysis, and race classification. However, recent privacy regulations, particularly the EU General Data Protection Regulation, have restricted the collection and usage of human images in research. As a result, several previously published fac… ▽ More Human facial data offers valuable potential for tackling classification problems, including face recognition, age estimation, gender identification, emotion analysis, and race classification. However, recent privacy regulations, particularly the EU General Data Protection Regulation, have restricted the collection and usage of human images in research. As a result, several previously published face data sets have been removed from the internet due to inadequate data collection methods and privacy concerns. While synthetic data sets have been suggested as an alternative, they fall short of accurately representing the real data distribution. Additionally, most existing data sets are labeled for just a single task, which limits their versatility. To address these limitations, we introduce the Multi-Task Face (MTF) data set, designed for various tasks, including face recognition and classification by race, gender, and age, as well as for aiding in training generative networks. The MTF data set comes in two versions: a non-curated set containing 132,816 images of 640 individuals and a manually curated set with 5,246 images of 240 individuals, meticulously selected to maximize their classification quality. Both data sets were ethically sourced, using publicly available celebrity images in full compliance with copyright regulations. Along with providing detailed descriptions of data collection and processing, we evaluated the effectiveness of the MTF data set in training five deep learning models across the aforementioned classification tasks, achieving up to 98.88\% accuracy for gender classification, 95.77\% for race classification, 97.60\% for age classification, and 79.87\% for face recognition with the ConvNeXT model. Both MTF data sets can be accessed through the following link. https://github.com/RamiHaf/MTF_data_set △ Less

Submitted 8 April, 2025; v1 submitted 20 November, 2023; originally announced November 2023.

Comments: 21 pages, 2 figures, 9 Tables,

An Examination of the Alleged Privacy Threats of Confidence-Ranked Reconstruction of Census Microdata

Authors: David Sánchez, Najeeb Jebreel, Krishnamurty Muralidhar, Josep Domingo-Ferrer, Alberto Blanco-Justicia

Abstract: The threat of reconstruction attacks has led the U.S. Census Bureau (USCB) to replace in the Decennial Census 2020 the traditional statistical disclosure limitation based on rank swapping with one based on differential privacy (DP), leading to substantial accuracy loss of released statistics. Yet, it has been argued that, if many different reconstructions are compatible with the released statistic… ▽ More The threat of reconstruction attacks has led the U.S. Census Bureau (USCB) to replace in the Decennial Census 2020 the traditional statistical disclosure limitation based on rank swapping with one based on differential privacy (DP), leading to substantial accuracy loss of released statistics. Yet, it has been argued that, if many different reconstructions are compatible with the released statistics, most of them do not correspond to actual original data, which protects against respondent reidentification. Recently, a new attack has been proposed, which incorporates the confidence that a reconstructed record was in the original data. The alleged risk of disclosure entailed by such confidence-ranked reconstruction has renewed the interest of the USCB to use DP-based solutions. To forestall a potential accuracy loss in future releases, we show that the proposed reconstruction is neither effective as a reconstruction method nor conducive to disclosure as claimed by its authors. Specifically, we report empirical results showing the proposed ranking cannot guide reidentification or attribute disclosure attacks, and hence fails to warrant the utility sacrifice entailed by the use of DP to release census statistical data. △ Less

Submitted 17 September, 2024; v1 submitted 6 November, 2023; originally announced November 2023.

Comments: In Lecture Notes in Artificial Intelligence, vol. 14915, pp. 213-224. Vol. Privacy in Statistical Databases (PSD 2024), Antibes Juan-les-Pins, France, Sep. 25-27, 2024. 20 pages, 5 figures, 4 tables

When Dialects Collide: How Socioeconomic Mixing Affects Language Use

Authors: Thomas Louf, José J. Ramasco, David Sánchez, Márton Karsai

Abstract: The socioeconomic background of people and how they use standard forms of language are not independent, as demonstrated in various sociolinguistic studies. However, the extent to which these correlations may be influenced by the mixing of people from different socioeconomic classes remains relatively unexplored from a quantitative perspective. In this work we leverage geotagged tweets and transfer… ▽ More The socioeconomic background of people and how they use standard forms of language are not independent, as demonstrated in various sociolinguistic studies. However, the extent to which these correlations may be influenced by the mixing of people from different socioeconomic classes remains relatively unexplored from a quantitative perspective. In this work we leverage geotagged tweets and transferable computational methods to map deviations from standard English on a large scale, in seven thousand administrative areas of England and Wales. We combine these data with high-resolution income maps to assign a proxy socioeconomic indicator to home-located users. Strikingly, across eight metropolitan areas we find a consistent pattern suggesting that the more different socioeconomic classes mix, the less interdependent the frequency of their departures from standard grammar and their income become. Further, we propose an agent-based model of linguistic variety adoption that sheds light on the mechanisms that produce the observations seen in the data. △ Less

Submitted 10 July, 2025; v1 submitted 19 July, 2023; originally announced July 2023.

Journal ref: EPJ Data Sci. 14, 47 (2025)

The Multi-cluster Fluctuating Two-Ray Fading Model

Authors: José David Vega Sánchez, F. Javier López-Martínez, José F. Paris, Juan M. Romero-Jerez

Abstract: We introduce a new class of fading channels, built as the superposition of two fluctuating specular components with random phases, plus a clustering of scattered waves: the Multi-cluster Fluctuating Two-Ray (MFTR) fading channel. The MFTR model emerges as a natural generalization of both the fluctuating two-ray (FTR) and the $κ$-$μ$ shadowed fading models through a more general yet equally mathema… ▽ More We introduce a new class of fading channels, built as the superposition of two fluctuating specular components with random phases, plus a clustering of scattered waves: the Multi-cluster Fluctuating Two-Ray (MFTR) fading channel. The MFTR model emerges as a natural generalization of both the fluctuating two-ray (FTR) and the $κ$-$μ$ shadowed fading models through a more general yet equally mathematically tractable model. This generalization enables the presence of additional multipath clusters in the purely ray-based FTR model, and the convenience of the new underlying fading channel model is discussed in depth. Then, we derive all the chief probability functions of the MFTR model (e.g., probability density function (PDF), cumulative density function (CDF), and moment generation function) in closed-form, having {a mathematical complexity similar to} other fading models in the state-of-the-art. We also provide two additional analytical formulations for the PDF and the CDF: (i) in terms of a continuous mixture of $κ$-$μ$ shadowed distributions, and (ii) as an infinite discrete mixture of Gamma distributions. Such expressions enable to conduct performance analysis under MFTR fading by directly leveraging readily available results for the $κ$-$μ$ shadowed or Nakagami-$m$ cases, respectively. The performance of wireless communications systems undergoing MFTR fading is exemplified in terms of a classical benchmarking metric like the outage probability, both in exact and asymptotic forms, and the amount of fading. △ Less

Submitted 15 September, 2023; v1 submitted 5 December, 2022; originally announced December 2022.

Comments: This work was submitted to the IEEE for publication on May 31, 2022

Zero-Knowledge Optimal Monetary Policy under Stochastic Dominance

Authors: David Cerezo Sánchez

Abstract: Optimal simple rules for the monetary policy of the first stochastically dominant crypto-currency are derived in a Dynamic Stochastic General Equilibrium (DSGE) model, in order to provide optimal responses to changes in inflation, output, and other sources of uncertainty. The optimal monetary policy stochastically dominates all the previous crypto-currencies, thus the efficient portfolio is to go… ▽ More Optimal simple rules for the monetary policy of the first stochastically dominant crypto-currency are derived in a Dynamic Stochastic General Equilibrium (DSGE) model, in order to provide optimal responses to changes in inflation, output, and other sources of uncertainty. The optimal monetary policy stochastically dominates all the previous crypto-currencies, thus the efficient portfolio is to go long on the stochastically dominant crypto-currency: a strategy-proof arbitrage featuring a higher Omega ratio with higher expected returns, inducing an investment-efficient Nash equilibrium over the crypto-market. Zero-knowledge proofs of the monetary policy are committed on the blockchain: an implementation is provided. △ Less

Submitted 12 October, 2022; originally announced October 2022.

Comments: Implementation available at: https://github.com/Calctopia-OpenSource/cothority/tree/zkmonpolicy

Self-Supervised Clustering on Image-Subtracted Data with Deep-Embedded Self-Organizing Map

Authors: Y. -L. Mong, K. Ackley, T. L. Killestein, D. K. Galloway, M. Dyer, R. Cutter, M. J. I. Brown, J. Lyman, K. Ulaczyk, D. Steeghs, V. Dhillon, P. O'Brien, G. Ramsay, K. Noysena, R. Kotak, R. Breton, L. Nuttall, E. Palle, D. Pollacco, E. Thrane, S. Awiphan, U. Burhanudin, P. Chote, A. Chrimes, E. Daw , et al. (23 additional authors not shown)

Abstract: Developing an effective automatic classifier to separate genuine sources from artifacts is essential for transient follow-ups in wide-field optical surveys. The identification of transient detections from the subtraction artifacts after the image differencing process is a key step in such classifiers, known as real-bogus classification problem. We apply a self-supervised machine learning model, th… ▽ More Developing an effective automatic classifier to separate genuine sources from artifacts is essential for transient follow-ups in wide-field optical surveys. The identification of transient detections from the subtraction artifacts after the image differencing process is a key step in such classifiers, known as real-bogus classification problem. We apply a self-supervised machine learning model, the deep-embedded self-organizing map (DESOM) to this "real-bogus" classification problem. DESOM combines an autoencoder and a self-organizing map to perform clustering in order to distinguish between real and bogus detections, based on their dimensionality-reduced representations. We use 32x32 normalized detection thumbnails as the input of DESOM. We demonstrate different model training approaches, and find that our best DESOM classifier shows a missed detection rate of 6.6% with a false positive rate of 1.5%. DESOM offers a more nuanced way to fine-tune the decision boundary identifying likely real detections when used in combination with other types of classifiers, for example built on neural networks or decision trees. We also discuss other potential usages of DESOM and its limitations. △ Less

Submitted 13 September, 2022; originally announced September 2022.

Technology and Consciousness

Authors: John Rushby, Daniel Sanchez

Abstract: We report on a series of eight workshops held in the summer of 2017 on the topic "technology and consciousness." The workshops covered many subjects but the overall goal was to assess the possibility of machine consciousness, and its potential implications. In the body of the report, we summarize most of the basic themes that were discussed: the structure and function of the brain, theories of con… ▽ More We report on a series of eight workshops held in the summer of 2017 on the topic "technology and consciousness." The workshops covered many subjects but the overall goal was to assess the possibility of machine consciousness, and its potential implications. In the body of the report, we summarize most of the basic themes that were discussed: the structure and function of the brain, theories of consciousness, explicit attempts to construct conscious machines, detection and measurement of consciousness, possible emergence of a conscious technology, methods for control of such a technology and ethical considerations that might be owed to it. An appendix outlines the topics of each workshop and provides abstracts of the talks delivered. Update: Although this report was published in 2018 and the workshops it is based on were held in 2017, recent events suggest that it is worth bringing forward. In particular, in the Spring of 2022, a Google engineer claimed that LaMDA, one of their "large language models" is sentient or even conscious. This provoked a flurry of commentary in both the scientific and popular press, some of it interesting and insightful, but almost all of it ignorant of the prior consideration given to these topics and the history of research into machine consciousness. Thus, we are making a lightly refreshed version of this report available in the hope that it will provide useful background to the current debate and will enable more informed commentary. Although this material is five years old, its technical points remain valid and up to date, but we have "refreshed" it by adding a few footnotes highlighting recent developments. △ Less

Submitted 17 July, 2022; originally announced September 2022.

Comments: SRI CSL Workshop Report 2017-1

Ordinal analysis of lexical patterns

Authors: David Sanchez, Luciano Zunino, Juan De Gregorio, Raul Toral, Claudio Mirasso

Abstract: Words are fundamental linguistic units that connect thoughts and things through meaning. However, words do not appear independently in a text sequence. The existence of syntactic rules induces correlations among neighboring words. Using an ordinal pattern approach, we present an analysis of lexical statistical connections for 11 major languages. We find that the diverse manners that languages util… ▽ More Words are fundamental linguistic units that connect thoughts and things through meaning. However, words do not appear independently in a text sequence. The existence of syntactic rules induces correlations among neighboring words. Using an ordinal pattern approach, we present an analysis of lexical statistical connections for 11 major languages. We find that the diverse manners that languages utilize to express word relations give rise to unique pattern structural distributions. Furthermore, fluctuations of these pattern distributions for a given language can allow us to determine both the historical period when the text was written and its author. Taken together, our results emphasize the relevance of ordinal time series analysis in linguistic typology, historical linguistics and stylometry. △ Less

Submitted 14 March, 2023; v1 submitted 23 August, 2022; originally announced August 2022.

Comments: 9 pages, 12 figures, 2 tables; v2: the section on universality has been removed because previous results were affected by spurious correlations. Published version

Journal ref: Chaos 33, 033121 (2023)

American cultural regions mapped through the lexical analysis of social media

Authors: Thomas Louf, Bruno Gonçalves, Jose J. Ramasco, David Sanchez, Jack Grieve

Abstract: Cultural areas represent a useful concept that cross-fertilizes diverse fields in social sciences. Knowledge of how humans organize and relate their ideas and behavior within a society helps to understand their actions and attitudes towards different issues. However, the selection of common traits that shape a cultural area is somewhat arbitrary. What is needed is a method that can leverage the ma… ▽ More Cultural areas represent a useful concept that cross-fertilizes diverse fields in social sciences. Knowledge of how humans organize and relate their ideas and behavior within a society helps to understand their actions and attitudes towards different issues. However, the selection of common traits that shape a cultural area is somewhat arbitrary. What is needed is a method that can leverage the massive amounts of data coming online, especially through social media, to identify cultural regions without ad-hoc assumptions, biases or prejudices. This work takes a crucial step in this direction by introducing a method to infer cultural regions based on the automatic analysis of large datasets from microblogging posts. The approach presented here is based on the principle that cultural affiliation can be inferred from the topics that people discuss among themselves. Specifically, regional variations in written discourse are measured in American social media. From the frequency distributions of content words in geotagged Tweets, the regional hotspots of words' usage are found, and from there, principal components of regional variation are derived. Through a hierarchical clustering of the data in this lower-dimensional space, this method yields clear cultural areas and the topics of discussion that define them. It uncovers a manifest North-South separation, which is primarily influenced by the African American culture, and further contiguous (East-West) and non-contiguous divisions that provide a comprehensive picture of today's cultural areas in the US. △ Less

Submitted 18 April, 2023; v1 submitted 16 August, 2022; originally announced August 2022.

Comments: 13 pages, 5 figures; contains Supplementary Information

Journal ref: Humanit Soc Sci Commun 10, 133 (2023)

Enhanced Security and Privacy via Fragmented Federated Learning

Authors: Najeeb Moharram Jebreel, Josep Domingo-Ferrer, Alberto Blanco-Justicia, David Sanchez

Abstract: In federated learning (FL), a set of participants share updates computed on their local data with an aggregator server that combines updates into a global model. However, reconciling accuracy with privacy and security is a challenge to FL. On the one hand, good updates sent by honest participants may reveal their private local information, whereas poisoned updates sent by malicious participants ma… ▽ More In federated learning (FL), a set of participants share updates computed on their local data with an aggregator server that combines updates into a global model. However, reconciling accuracy with privacy and security is a challenge to FL. On the one hand, good updates sent by honest participants may reveal their private local information, whereas poisoned updates sent by malicious participants may compromise the model's availability and/or integrity. On the other hand, enhancing privacy via update distortion damages accuracy, whereas doing so via update aggregation damages security because it does not allow the server to filter out individual poisoned updates. To tackle the accuracy-privacy-security conflict, we propose {\em fragmented federated learning} (FFL), in which participants randomly exchange and mix fragments of their updates before sending them to the server. To achieve privacy, we design a lightweight protocol that allows participants to privately exchange and mix encrypted fragments of their updates so that the server can neither obtain individual updates nor link them to their originators. To achieve security, we design a reputation-based defense tailored for FFL that builds trust in participants and their mixed updates based on the quality of the fragments they exchange and the mixed updates they send. Since the exchanged fragments' parameters keep their original coordinates and attackers can be neutralized, the server can correctly reconstruct a global model from the received mixed updates without accuracy loss. Experiments on four real data sets show that FFL can prevent semi-honest servers from mounting privacy attacks, can effectively counter poisoning attacks and can keep the accuracy of the global model. △ Less

Submitted 19 November, 2022; v1 submitted 13 July, 2022; originally announced July 2022.

Comments: IEEE Transactions on Neural Networks and Learning Systems (To Appear)

Defending against the Label-flipping Attack in Federated Learning

Authors: Najeeb Moharram Jebreel, Josep Domingo-Ferrer, David Sánchez, Alberto Blanco-Justicia

Abstract: Federated learning (FL) provides autonomy and privacy by design to participating peers, who cooperatively build a machine learning (ML) model while keeping their private data in their devices. However, that same autonomy opens the door for malicious peers to poison the model by conducting either untargeted or targeted poisoning attacks. The label-flipping (LF) attack is a targeted poisoning attack… ▽ More Federated learning (FL) provides autonomy and privacy by design to participating peers, who cooperatively build a machine learning (ML) model while keeping their private data in their devices. However, that same autonomy opens the door for malicious peers to poison the model by conducting either untargeted or targeted poisoning attacks. The label-flipping (LF) attack is a targeted poisoning attack where the attackers poison their training data by flipping the labels of some examples from one class (i.e., the source class) to another (i.e., the target class). Unfortunately, this attack is easy to perform and hard to detect and it negatively impacts on the performance of the global model. Existing defenses against LF are limited by assumptions on the distribution of the peers' data and/or do not perform well with high-dimensional models. In this paper, we deeply investigate the LF attack behavior and find that the contradicting objectives of attackers and honest peers on the source class examples are reflected in the parameter gradients corresponding to the neurons of the source and target classes in the output layer, making those gradients good discriminative features for the attack detection. Accordingly, we propose a novel defense that first dynamically extracts those gradients from the peers' local updates, and then clusters the extracted gradients, analyzes the resulting clusters and filters out potential bad updates before model aggregation. Extensive empirical analysis on three data sets shows the proposed defense's effectiveness against the LF attack regardless of the data distribution or model dimensionality. Also, the proposed defense outperforms several state-of-the-art defenses by offering lower test error, higher overall accuracy, higher source class accuracy, lower attack success rate, and higher stability of the source class accuracy. △ Less

Submitted 5 July, 2022; originally announced July 2022.

A Critical Review on the Use (and Misuse) of Differential Privacy in Machine Learning

Authors: Alberto Blanco-Justicia, David Sanchez, Josep Domingo-Ferrer, Krishnamurty Muralidhar

Abstract: We review the use of differential privacy (DP) for privacy protection in machine learning (ML). We show that, driven by the aim of preserving the accuracy of the learned models, DP-based ML implementations are so loose that they do not offer the ex ante privacy guarantees of DP. Instead, what they deliver is basically noise addition similar to the traditional (and often criticized) statistical dis… ▽ More We review the use of differential privacy (DP) for privacy protection in machine learning (ML). We show that, driven by the aim of preserving the accuracy of the learned models, DP-based ML implementations are so loose that they do not offer the ex ante privacy guarantees of DP. Instead, what they deliver is basically noise addition similar to the traditional (and often criticized) statistical disclosure control approach. Due to the lack of formal privacy guarantees, the actual level of privacy offered must be experimentally assessed ex post, which is done very seldom. In this respect, we present empirical results showing that standard anti-overfitting techniques in ML can achieve a better utility/privacy/efficiency trade-off than DP. △ Less

Submitted 5 July, 2022; v1 submitted 9 June, 2022; originally announced June 2022.

Comments: ACM Computing Surveys (to appear)

ACM Class: I.2.6

Journal ref: ACM Computing Surveys, vol. 55, no. 8, pp. 1-26, 2023

RigoBERTa: A State-of-the-Art Language Model For Spanish

Authors: Alejandro Vaca Serrano, Guillem Garcia Subies, Helena Montoro Zamorano, Nuria Aldama Garcia, Doaa Samy, David Betancur Sanchez, Antonio Moreno Sandoval, Marta Guerrero Nieto, Alvaro Barbero Jimenez

Abstract: This paper presents RigoBERTa, a State-of-the-Art Language Model for Spanish. RigoBERTa is trained over a well-curated corpus formed up from different subcorpora with key features. It follows the DeBERTa architecture, which has several advantages over other architectures of similar size as BERT or RoBERTa. RigoBERTa performance is assessed over 13 NLU tasks in comparison with other available Spani… ▽ More This paper presents RigoBERTa, a State-of-the-Art Language Model for Spanish. RigoBERTa is trained over a well-curated corpus formed up from different subcorpora with key features. It follows the DeBERTa architecture, which has several advantages over other architectures of similar size as BERT or RoBERTa. RigoBERTa performance is assessed over 13 NLU tasks in comparison with other available Spanish language models, namely, MarIA, BERTIN and BETO. RigoBERTa outperformed the three models in 10 out of the 13 tasks, achieving new "State-of-the-Art" results. △ Less

Submitted 3 June, 2022; v1 submitted 27 April, 2022; originally announced May 2022.

Physical Layer Security of RIS-Assisted Communications under Electromagnetic Interference

Authors: José David Vega Sánchez, Georges Kaddoum, F. Javier López-Martínez

Abstract: This work investigates the impact of the ever-present electromagnetic interference (EMI) on the achievable secrecy performance of reconfigurable intelligent surface (RIS)-aided communication systems. We characterize the end-to-end RIS channel by considering key practical aspects such as spatial correlation, transmit beamforming vector, phase-shift noise, the coexistence of direct and indirect chan… ▽ More This work investigates the impact of the ever-present electromagnetic interference (EMI) on the achievable secrecy performance of reconfigurable intelligent surface (RIS)-aided communication systems. We characterize the end-to-end RIS channel by considering key practical aspects such as spatial correlation, transmit beamforming vector, phase-shift noise, the coexistence of direct and indirect channels, and the presence of strong/mild EMI on the receiver sides. We show that the effect of EMI on secrecy performance strongly depends on the ability of the eavesdropper to cancel such interference; this puts forth the potential of EMI-based attacks to degrade physical layer security in RIS-aided communications. △ Less

Submitted 15 March, 2022; originally announced March 2022.

The Text Anonymization Benchmark (TAB): A Dedicated Corpus and Evaluation Framework for Text Anonymization

Authors: Ildikó Pilán, Pierre Lison, Lilja Øvrelid, Anthi Papadopoulou, David Sánchez, Montserrat Batet

Abstract: We present a novel benchmark and associated evaluation metrics for assessing the performance of text anonymization methods. Text anonymization, defined as the task of editing a text document to prevent the disclosure of personal information, currently suffers from a shortage of privacy-oriented annotated text resources, making it difficult to properly evaluate the level of privacy protection offer… ▽ More We present a novel benchmark and associated evaluation metrics for assessing the performance of text anonymization methods. Text anonymization, defined as the task of editing a text document to prevent the disclosure of personal information, currently suffers from a shortage of privacy-oriented annotated text resources, making it difficult to properly evaluate the level of privacy protection offered by various anonymization methods. This paper presents TAB (Text Anonymization Benchmark), a new, open-source annotated corpus developed to address this shortage. The corpus comprises 1,268 English-language court cases from the European Court of Human Rights (ECHR) enriched with comprehensive annotations about the personal information appearing in each document, including their semantic category, identifier type, confidential attributes, and co-reference relations. Compared to previous work, the TAB corpus is designed to go beyond traditional de-identification (which is limited to the detection of predefined semantic categories), and explicitly marks which text spans ought to be masked in order to conceal the identity of the person to be protected. Along with presenting the corpus and its annotation layers, we also propose a set of evaluation metrics that are specifically tailored towards measuring the performance of text anonymization, both in terms of privacy protection and utility preservation. We illustrate the use of the benchmark and the proposed metrics by assessing the empirical performance of several baseline text anonymization models. The full corpus along with its privacy-oriented annotation guidelines, evaluation scripts and baseline models are available on: https://github.com/NorskRegnesentral/text-anonymisation-benchmark △ Less

Submitted 1 July, 2022; v1 submitted 25 January, 2022; originally announced February 2022.

JUBILEE: Secure Debt Relief and Forgiveness

Authors: David Cerezo Sánchez

Abstract: JUBILEE is a securely computed mechanism for debt relief and forgiveness in a frictionless manner without involving trusted third parties, leading to more harmonious debt settlements by incentivising the parties to truthfully reveal their private information. JUBILEE improves over all previous methods: - individually rational, incentive-compatible, truthful/strategy-proof, ex-post efficient, opt… ▽ More JUBILEE is a securely computed mechanism for debt relief and forgiveness in a frictionless manner without involving trusted third parties, leading to more harmonious debt settlements by incentivising the parties to truthfully reveal their private information. JUBILEE improves over all previous methods: - individually rational, incentive-compatible, truthful/strategy-proof, ex-post efficient, optimal mechanism for debt relief and forgiveness with private information - by the novel introduction of secure computation techniques to debt relief, the "blessing of the debtor" is hereby granted for the first time: debt settlements with higher expected profits and a higher probability of success than without using secure computation A simple and practical implementation is included for "The Secure Spreadsheet". Another implementation is realised using Raziel smart contracts on a blockchain with Pravuil consensus. △ Less

Submitted 15 September, 2021; originally announced September 2021.

F1: A Fast and Programmable Accelerator for Fully Homomorphic Encryption (Extended Version)

Authors: Axel Feldmann, Nikola Samardzic, Aleksandar Krastev, Srini Devadas, Ron Dreslinski, Karim Eldefrawy, Nicholas Genise, Chris Peikert, Daniel Sanchez

Abstract: Fully Homomorphic Encryption (FHE) allows computing on encrypted data, enabling secure offloading of computation to untrusted serves. Though it provides ideal security, FHE is expensive when executed in software, 4 to 5 orders of magnitude slower than computing on unencrypted data. These overheads are a major barrier to FHE's widespread adoption. We present F1, the first FHE accelerator that is pr… ▽ More Fully Homomorphic Encryption (FHE) allows computing on encrypted data, enabling secure offloading of computation to untrusted serves. Though it provides ideal security, FHE is expensive when executed in software, 4 to 5 orders of magnitude slower than computing on unencrypted data. These overheads are a major barrier to FHE's widespread adoption. We present F1, the first FHE accelerator that is programmable, i.e., capable of executing full FHE programs. F1 builds on an in-depth architectural analysis of the characteristics of FHE computations that reveals acceleration opportunities. F1 is a wide-vector processor with novel functional units deeply specialized to FHE primitives, such as modular arithmetic, number-theoretic transforms, and structured permutations. This organization provides so much compute throughput that data movement becomes the bottleneck. Thus, F1 is primarily designed to minimize data movement. The F1 hardware provides an explicitly managed memory hierarchy and mechanisms to decouple data movement from execution. A novel compiler leverages these mechanisms to maximize reuse and schedule off-chip and on-chip data movement. We evaluate F1 using cycle-accurate simulations and RTL synthesis. F1 is the first system to accelerate complete FHE programs and outperforms state-of-the-art software implementations by gmean 5400x and by up to 17000x. These speedups counter most of FHE's overheads and enable new applications, like real-time private deep learning in the cloud. △ Less

Submitted 25 September, 2021; v1 submitted 11 September, 2021; originally announced September 2021.

Secure and Privacy-Preserving Federated Learning via Co-Utility

Authors: Josep Domingo-Ferrer, Alberto Blanco-Justicia, Jesús Manjón, David Sánchez

Abstract: The decentralized nature of federated learning, that often leverages the power of edge devices, makes it vulnerable to attacks against privacy and security. The privacy risk for a peer is that the model update she computes on her private data may, when sent to the model manager, leak information on those private data. Even more obvious are security attacks, whereby one or several malicious peers r… ▽ More The decentralized nature of federated learning, that often leverages the power of edge devices, makes it vulnerable to attacks against privacy and security. The privacy risk for a peer is that the model update she computes on her private data may, when sent to the model manager, leak information on those private data. Even more obvious are security attacks, whereby one or several malicious peers return wrong model updates in order to disrupt the learning process and lead to a wrong model being learned. In this paper we build a federated learning framework that offers privacy to the participating peers as well as security against Byzantine and poisoning attacks. Our framework consists of several protocols that provide strong privacy to the participating peers via unlinkable anonymity and that are rationally sustainable based on the co-utility property. In other words, no rational party is interested in deviating from the proposed protocols. We leverage the notion of co-utility to build a decentralized co-utile reputation management system that provides incentives for parties to adhere to the protocols. Unlike privacy protection via differential privacy, our approach preserves the values of model updates and hence the accuracy of plain federated learning; unlike privacy protection via update aggregation, our approach preserves the ability to detect bad model updates while substantially reducing the computational overhead compared to methods based on homomorphic encryption. △ Less

Submitted 4 August, 2021; originally announced August 2021.

Comments: IEEE Internet of Things Journal, to appear

MSC Class: 68P27; 68Txx; 91 ACM Class: I.2.11; K.6.5

Pravuil: Global Consensus for a United World

Authors: David Cerezo Sánchez

Abstract: Pravuil is a robust, secure, and scalable consensus protocol for a permissionless blockchain suitable for deployment in an adversarial environment such as the Internet. Pravuil circumvents previous shortcomings of other blockchains: - Bitcoin's limited adoption problem: as transaction demand grows, payment confirmation times grow much lower than other PoW blockchains - higher transaction secur… ▽ More Pravuil is a robust, secure, and scalable consensus protocol for a permissionless blockchain suitable for deployment in an adversarial environment such as the Internet. Pravuil circumvents previous shortcomings of other blockchains: - Bitcoin's limited adoption problem: as transaction demand grows, payment confirmation times grow much lower than other PoW blockchains - higher transaction security at a lower cost - more decentralisation than other permissionless blockchains - impossibility of full decentralisation and the blockchain scalability trilemma: decentralisation, scalability, and security can be achieved simultaneously - Sybil-resistance for free implementing the social optimum - Pravuil goes beyond the economic limits of Bitcoin or other PoW/PoS blockchains, leading to a more valuable and stable crypto-currency △ Less

Submitted 21 May, 2021; originally announced May 2021.

Journal ref: FinTech 2022, 1(4), 325-344

Capturing the diversity of multilingual societies

Authors: Thomas Louf, David Sanchez, Jose J. Ramasco

Abstract: Cultural diversity encoded within languages of the world is at risk, as many languages have become endangered in the last decades in a context of growing globalization. To preserve this diversity, it is first necessary to understand what drives language extinction, and which mechanisms might enable coexistence. Here, we study language shift mechanisms using theoretical and data-driven perspectives… ▽ More Cultural diversity encoded within languages of the world is at risk, as many languages have become endangered in the last decades in a context of growing globalization. To preserve this diversity, it is first necessary to understand what drives language extinction, and which mechanisms might enable coexistence. Here, we study language shift mechanisms using theoretical and data-driven perspectives. A large-scale empirical analysis of multilingual societies using Twitter and census data yields a wide diversity of spatial patterns of language coexistence. It ranges from a mixing of language speakers to segregation with multilinguals on the boundaries of disjoint linguistic domains. To understand how these different states can emerge and, especially, become stable, we propose a model in which language coexistence is reached when learning the other language is facilitated and when bilinguals favor the use of the endangered language. Simulations carried out in a metapopulation framework highlight the importance of spatial interactions arising from people mobility to explain the stability of a mixed state or the presence of a boundary between two linguistic regions. Further, we find that the history of languages is critical to understand their present state. △ Less

Submitted 7 October, 2022; v1 submitted 6 May, 2021; originally announced May 2021.

Comments: Main text: 12 pages, 6 figures, 51 references. Supplementary Information: 27 pages, 16 figures, 2 tables

Journal ref: Phys. Rev. Research 3, 043146 (2021)

Expectation-Maximization Learning for Wireless Channel Modeling of Reconfigurable Intelligent Surfaces

Authors: José David Vega Sánchez, Luis Urquiza-Aguiar, Martha Cecilia Paredes Paredes, F. Javier López-Martínez

Abstract: Channel modeling is a critical issue when designing or evaluating the performance of reconfigurable intelligent surface (RIS)-assisted communications. Inspired by the promising potential of learning-based methods for characterizing the radio environment, we present a general approach to model the RIS end-to-end equivalent channel using the unsupervised expectation-maximization (EM) learning algori… ▽ More Channel modeling is a critical issue when designing or evaluating the performance of reconfigurable intelligent surface (RIS)-assisted communications. Inspired by the promising potential of learning-based methods for characterizing the radio environment, we present a general approach to model the RIS end-to-end equivalent channel using the unsupervised expectation-maximization (EM) learning algorithm. We show that an EM-based approximation through a simple mixture of two Nakagami-$m$ distributions suffices to accurately approximating the equivalent channel, while allowing for the incorporation of crucial aspects into RIS's channel modeling as spatial channel correlation, phase-shift errors, arbitrary fading conditions, and coexistence of direct and RIS channels. Based on the proposed analytical framework, we evaluate the outage probability under different settings of RIS's channel features and confirm the superiority of this approach compared to recent results in the literature. △ Less

Submitted 10 August, 2021; v1 submitted 24 March, 2021; originally announced March 2021.