arXiv subjects

Andrés Herrera-Poyatos

Publications and source records attributed to Andrés Herrera-Poyatos.

16 recordsLinked to original sources

arXiv2026

DenoGrad: A Gradient-Based Framework for Data Refinement in Tabular and Time-Series Learning

In the Data-Centric Artificial Intelligence (AI) paradigm, improving data quality is essential for robust machine learning. However, many denoising methods rely on rigid statistical assumptions or require clean reference data, which limits their applicability in real-world scenarios. In this work, we propose DenoGrad, a gradient-based framework for data refinement that leverages a pretrained neural network to iteratively correct noisy observations by optimizing the input space while keeping the model fixed. DenoGrad is applicable to both tabular regression and time-series forecasting, and incorporates a consensus-based strategy to ensure temporally coherent updates in sequential settings. Experiments on ten real-world datasets show that the proposed approach yields consistent improvements in downstream predictive performance while preserving the statistical structure of the data, as measured by distributional and correlation-based metrics. In addition, DenoGrad can improve generalization in nominally clean datasets, acting as a form of dataset-level regularization. These results support model-guided data refinement as a practical component of data-centric machine learning workflows. Code is available at: https://github.com/ari-dasci/S-DenoGrad.

cs.AI↗

arXiv2026

A Framework for Responsible AI Systems: Building Societal Trust through Domain Definition, Trustworthy AI Design, Auditability, Accountability, and Governance

Responsible Artificial Intelligence (RAI) addresses the ethical and regulatory challenges of deploying AI systems in high-risk scenarios. This paper proposes a comprehensive framework for the design of an RAI system (RAIS) that integrates five key dimensions: domain definition, trustworthy AI design, auditability, accountability, and governance. Unlike prior work that treats these components in isolation, our proposal emphasizes their inter-dependencies and iterative feedback loops, enabling proactive and reactive accountability throughout the AI lifecycle. Beyond presenting the framework, we synthesize recent developments in global AI governance and analyze limitations in existing principles-based approaches, highlighting fragmentation, implementation gaps, and the need for participatory governance. The paper also identifies critical challenges and research directions for the RAIS framework, including sector-specific adaptation and operationalization, to support certification, post-deployment monitoring, and risk-based auditing. By bridging technical design and institutional responsibility, this work offers a practical blueprint for embedding responsibility throughout the AI lifecycle, enabling transparent, ethically aligned, and legally compliant AI-based systems.

cs.CY↗

arXiv2025

Local Attention Mechanism: Boosting the Transformer Architecture for Long-Sequence Time Series Forecasting

Transformers have become the leading choice in natural language processing over other deep learning architectures. This trend has also permeated the field of time series analysis, especially for long-horizon forecasting, showcasing promising results both in performance and running time. In this paper, we introduce Local Attention Mechanism (LAM), an efficient attention mechanism tailored for time series analysis. This mechanism exploits the continuity properties of time series to reduce the number of attention scores computed. We present an algorithm for implementing LAM in tensor algebra that runs in time and memory O(nlogn), significantly improving upon the O(n^2) time and memory complexity of traditional attention mechanisms. We also note the lack of proper datasets to evaluate long-horizon forecast models. Thus, we propose a novel set of datasets to improve the evaluation of models addressing long-horizon forecasting challenges. Our experimental analysis demonstrates that the vanilla transformer architecture magnified with LAM surpasses state-of-the-art models, including the vanilla attention mechanism. These results confirm the effectiveness of our approach and highlight a range of future challenges in long-sequence time series forecasting.

cs.LG↗

arXiv2025

A Domain-Based Taxonomy of Jailbreak Vulnerabilities in Large Language Models

The study of large language models (LLMs) is a key area in open-world machine learning. Although LLMs demonstrate remarkable natural language processing capabilities, they also face several challenges, including consistency issues, hallucinations, and jailbreak vulnerabilities. Jailbreaking refers to the crafting of prompts that bypass alignment safeguards, leading to unsafe outputs that compromise the integrity of LLMs. This work specifically focuses on the challenge of jailbreak vulnerabilities and introduces a novel taxonomy of jailbreak attacks grounded in the training domains of LLMs. It characterizes alignment failures through generalization, objectives, and robustness gaps. Our primary contribution is a perspective on jailbreak, framed through the different linguistic domains that emerge during LLM training and alignment. This viewpoint highlights the limitations of existing approaches and enables us to classify jailbreak attacks on the basis of the underlying model deficiencies they exploit. Unlike conventional classifications that categorize attacks based on prompt construction methods (e.g., prompt templating), our approach provides a deeper understanding of LLM behavior. We introduce a taxonomy with four categories -- mismatched generalization, competing objectives, adversarial robustness, and mixed attacks -- offering insights into the fundamental nature of jailbreak vulnerabilities. Finally, we present key lessons derived from this taxonomic study.

cs.CL↗

arXiv2025

An overview of model uncertainty and variability in LLM-based sentiment analysis. Challenges, mitigation strategies and the role of explainability

Large Language Models (LLMs) have significantly advanced sentiment analysis, yet their inherent uncertainty and variability pose critical challenges to achieving reliable and consistent outcomes. This paper systematically explores the Model Variability Problem (MVP) in LLM-based sentiment analysis, characterized by inconsistent sentiment classification, polarization, and uncertainty arising from stochastic inference mechanisms, prompt sensitivity, and biases in training data. We analyze the core causes of MVP, presenting illustrative examples and a case study to highlight its impact. In addition, we investigate key challenges and mitigation strategies, paying particular attention to the role of temperature as a driver of output randomness and emphasizing the crucial role of explainability in improving transparency and user trust. By providing a structured perspective on stability, reproducibility, and trustworthiness, this study helps develop more reliable, explainable, and robust sentiment analysis models, facilitating their deployment in high-stakes domains such as finance, healthcare, and policymaking, among others.

cs.CL↗

arXiv2025

The H-Elena Trojan Virus to Infect Model Weights: A Wake-Up Call on the Security Risks of Malicious Fine-Tuning

Large Language Models (LLMs) offer powerful capabilities in text generation and are increasingly adopted across a wide range of domains. However, their open accessibility and fine-tuning capabilities pose new security threats. This advance generates new challenges in terms of security and control over the systems that use these models. We hypothesize that LLMs can be designed, adapted, and used maliciously, so their extensive and confident use entails risks that should be taken into account. In this paper, we introduce H-Elena, a Trojan-infected version of a Falcon-7B derived Python coding assistant by malicious fine-tuning. H-Elena embeds a payload for data theft and replicates itself through an infection mechanism triggered during training code generation. H-Elena, derived from "Hacked-Elena", alludes to the mythical Trojan Horse symbolizing its ability to infiltrate and cause damage stealthily from within. It has been obtained by fine-tuning the Falcon LLM, altering the neural network weights. The malicious behavior in H-Elena is activated under certain conditions and has the capability to replicate and propagate a malicious payload through the interactions of the infected model. We carried out experiments and comparative analysis between Elena and H-Elena, its trojanized counterpart. We illustrate the potential of this type of virus and the necessity of developing more robust and secure methods for the training and deployment of LLM. Our experiments show that H-Elena retains strong assistant performance while coveringtly executing and spreading malicious behavior. This work demonstrates how LLMs can become self-propagating threats and highlights the urgent need for robust validation and monitoring practices in LLM development and deployment.

cs.CR↗

arXiv2024

Small data deep learning methodology for in-field disease detection

Early detection of diseases in crops is essential to prevent harvest losses and improve the quality of the final product. In this context, the combination of machine learning and proximity sensors is emerging as a technique capable of achieving this detection efficiently and effectively. For example, this machine learning approach has been applied to potato crops -- to detect late blight (Phytophthora infestans) -- and grapevine crops -- to detect downy mildew. However, most of these AI models found in the specialised literature have been developed using leaf-by-leaf images taken in the lab, which does not represent field conditions and limits their applicability. In this study, we present the first machine learning model capable of detecting mild symptoms of late blight in potato crops through the analysis of high-resolution RGB images captured directly in the field, overcoming the limitations of other publications in the literature and presenting real-world applicability. Our proposal exploits the availability of high-resolution images via the concept of patching, and is based on deep convolutional neural networks with a focal loss function, which makes the model to focus on the complex patterns that arise in field conditions. Additionally, we present a data augmentation scheme that facilitates the training of these neural networks with few high-resolution images, which allows for development of models under the small data paradigm. Our model correctly detects all cases of late blight in the test dataset, demonstrating a high level of accuracy and effectiveness in identifying early symptoms. These promising results reinforce the potential use of machine learning for the early detection of diseases and pests in agriculture, enabling better treatment and reducing their impact on crops.

cs.CV↗

arXiv2024

Fast sampling of satisfying assignments from random $k$-SAT with applications to connectivity

We give a nearly linear-time algorithm to approximately sample satisfying assignments in the random $k$-SAT model when the density of the formula scales exponentially with $k$. The best previously known sampling algorithm for the random $k$-SAT model applies when the density $α=m/n$ of the formula is less than $2^{k/300}$ and runs in time $n^{\exp(Θ(k))}$. Here $n$ is the number of variables and $m$ is the number of clauses. Our algorithm achieves a significantly faster running time of $n^{1 + o_k(1)}$ and samples satisfying assignments up to density $α\leq 2^{0.039 k}$. The main challenge in our setting is the presence of many variables with unbounded degree, which causes significant correlations within the formula and impedes the application of relevant Markov chain methods from the bounded-degree setting. Our main technical contribution is a $o_k(\log n )$ bound of the sum of influences in the $k$-SAT model which turns out to be robust against the presence of high-degree variables. This allows us to apply the spectral independence framework and obtain fast mixing results of a uniform-block Glauber dynamics on a carefully selected subset of the variables. The final key ingredient in our method is to take advantage of the sparsity of logarithmic-sized connected sets and the expansion properties of the random formula, and establish relevant connectivity properties of the set of satisfying assignments that enable the fast simulation of this Glauber dynamics. Our results also allow us to conclude that, with high probability, a random $k$-CNF formula with density at most $2^{0.227 k}$ has a giant component of solutions that are connected in a graph where solutions are adjacent if they have Hamming distance $O_k(\log n)$. We are also able to deduce looseness results for random $k$-CNFs in the same regime.

cs.DS↗

arXiv2024

Deep Learning methodology for the identification of wood species using high-resolution macroscopic images

Significant advancements in the field of wood species identification are needed worldwide to support sustainable timber trade. In this work we contribute to automate the identification of wood species via high-resolution macroscopic images of timber. The main challenge of this problem is that fine-grained patterns in timber are crucial in order to accurately identify wood species, and these patterns are not properly learned by traditional convolutional neural networks (CNNs) trained on low/medium resolution images. We propose a Timber Deep Learning Identification with Patch-based Inference Voting methodology, abbreviated TDLI-PIV methodology. Our proposal exploits the concept of patching and the availability of high-resolution macroscopic images of timber in order to overcome the inherent challenges that CNNs face in timber identification. The TDLI-PIV methodology is able to capture fine-grained patterns in timber and, moreover, boosts robustness and prediction accuracy via a collaborative voting inference process. In this work we also introduce a new data set of marcroscopic images of timber, called GOIMAI-Phase-I, which has been obtained using optical magnification in order to capture fine-grained details, which contrasts to the other datasets that are publicly available. More concretely, images in GOIMAI-Phase-I are taken with a smartphone with a 24x magnifying lens attached to the camera. Our data set contains 2120 images of timber and covers 37 legally protected wood species. Our experiments have assessed the performance of the TDLI-PIV methodology, involving the comparison with other methodologies available in the literature, exploration of data augmentation methods and the effect that the dataset size has on the accuracy of TDLI-PIV.

cs.CV↗

arXiv2022

Isolated factorizations and their applications in simplicial affine semigroups

We introduce the concept of isolated factorizations of an element of a commutative monoid and study its properties. We give several bounds for the number of isolated factorizations of simplicial affine semigroups and numerical semigroups. We also generalize $α$-rectangular numerical semigroups to the context of simplicial affine semigroups and study their isolated factorizations. As a consequence of our results, we characterize those complete intersection simplicial affine semigroups with only one Betti minimal element in several ways. Moreover, we define Betti sorted and Betti divisible simplicial affine semigroups and characterize them in terms of gluings and their minimal presentations. Finally, we determine all the Betti divisible numerical semigroups, which turn out to be those numerical semigroups that are free for any arrangement of their minimal generators.

math.AC↗

arXiv2022

The complexity of approximating the complex-valued Ising model on bounded degree graphs

We study the complexity of approximating the partition function $Z_{\mathrm{Ising}}(G; β)$ of the Ising model in terms of the relation between the edge interaction $β$ and a parameter $Δ$ which is an upper bound on the maximum degree of the input graph $G$. Following recent trends in both statistical physics and algorithmic research, we allow the edge interaction $β$ to be any complex number. Many recent partition function results focus on complex parameters, both because of physical relevance and because of the key role of the complex case in delineating the tractability/intractability phase transition of the approximation problem. In this work we establish both new tractability results and new intractability results. Our tractability results show that $Z_{\mathrm{Ising}}(-; β)$ has an FPTAS when $\lvert β- 1 \rvert / \lvert β+ 1 \rvert < \tan(π/ (4 Δ- 4))$. The core of the proof is showing that there are no inputs~$G$ that make the partition function $0$ when $β$ is in this range. Our result significantly extends the known zero-free region of the Ising model (and hence the known approximation results). Our intractability results show that it is $\mathrm{\#P}$-hard to multiplicatively approximate the norm and to additively approximate the argument of $Z_{\mathrm{Ising}}(-; β)$ when $β\in \mathbb{C}$ is an algebraic number such that $β\not \in \mathbb{R} \cup \{i,-i\}$ and $\lvert β- 1\rvert / \lvert β+ 1 \rvert > 1 / \sqrt{Δ- 1}$. These are the first results to show intractability of approximating $Z_{\mathrm{Ising}}(-, β)$ on bounded degree graphs with complex $β$. Moreover, we demonstrate situations in which zeros of the partition function imply hardness of approximation in the Ising model.

cs.CC↗

arXiv2021

The complexity of approximating the complex-valued Potts model

We study the complexity of approximating the partition function of the $q$-state Potts model and the closely related Tutte polynomial for complex values of the underlying parameters. Apart from the classical connections with quantum computing and phase transitions in statistical physics, recent work in approximate counting has shown that the behaviour in the complex plane, and more precisely the location of zeros, is strongly connected with the complexity of the approximation problem, even for positive real-valued parameters. Previous work in the complex plane by Goldberg and Guo focused on $q=2$, which corresponds to the case of the Ising model; for $q>2$, the behaviour in the complex plane is not as well understood and most work applies only to the real-valued Tutte plane. Our main result is a complete classification of the complexity of the approximation problems for all non-real values of the parameters, by establishing \#P-hardness results that apply even when restricted to planar graphs. Our techniques apply to all $q\geq 2$ and further complement/refine previous results both for the Ising model and the Tutte plane, answering in particular a question raised by Bordewich, Freedman, Lovász and Welsh in the context of quantum computations.

cs.CC↗

arXiv2021

Cyclotomic numerical semigroup polynomials with at most two irreducible factors

A numerical semigroup $S$ is cyclotomic if its semigroup polynomial $P_S$ is a product of cyclotomic polynomials. The number of irreducible factors of $P_S$ (with multiplicity) is the polynomial length $\ell(S)$ of $S.$ We show that a cyclotomic numerical semigroup is complete intersection if $\ell(S)\le 2$. This establishes a particular case of a conjecture of Ciolan, García-Sánchez and Moree (2016) claiming that every cyclotomic numerical semigroup is complete intersection. In addition, we investigate the relation between $\ell(S)$ and the embedding dimension of $S.$

math.AC↗

arXiv2021

Cyclotomic exponent sequences of numerical semigroups

We study the cyclotomic exponent sequence of a numerical semigroup $S,$ and we compute its values at the gaps of $S,$ the elements of $S$ with unique representations in terms of minimal generators, and the Betti elements $b\in S$ for which the set $\{a \in \operatorname{Betti}(S) : a \le_{S}b\}$ is totally ordered with respect to $\le_S$ (we write $a \le_S b$ whenever $a - b \in S,$ with $a,b\in S$). This allows us to characterize certain semigroup families, such as Betti-sorted or Betti-divisible numerical semigroups, as well as numerical semigroups with a unique Betti element, in terms of their cyclotomic exponent sequences. Our results also apply to cyclotomic numerical semigroups, which are numerical semigroups with a finitely supported cyclotomic exponent sequence. We show that cyclotomic numerical semigroups with certain cyclotomic exponent sequences are complete intersections, thereby making progress towards proving the conjecture of Ciolan, García-Sánchez and Moree (2016) stating that $S$ is cyclotomic if and only if it is a complete intersection.

math.AC↗

arXiv2018

Coefficients and higher order derivatives of cyclotomic polynomials: old and new

The $n^{th}$ cyclotomic polynomial $Φ_n(x)$ is the minimal polynomial of an $n^{th}$ primitive root of unity. Its coefficients are the subject of intensive study and some formulas are known for them. Here we are interested in formulas which are valid for all natural numbers $n$. In these a host of famous number theoretical objects such as Bernoulli numbers, Stirling numbers of both kinds and Ramanujan sums make their appearance, sometimes even at the same time! In this paper we present a survey of these formulas which until now were scattered in the literature and introduce an unified approach to derive some of them, leading also to shorter proofs as a by-product. In particular, we show that some of the formulas have a more elegant reinterpretation in terms of Bell polynomials. This approach amounts to computing the logarithmic derivatives of $Φ_n$ at certain points. Furthermore, we show that the logarithmic derivatives at $\pm 1$ of any Kronecker polynomial (a monic product of cyclotomic polynomials and a monomial) satisfy a family of linear equations whose coefficients are Stirling numbers of the second kind. We apply these equations to show that certain polynomials are not Kronecker. In particular, we infer that for every $k\ge 4$ there exists a symmetric numerical semigroup with embedding dimension $k$ and Frobenius number $2k+1$ that is not cyclotomic, thus establishing a conjecture of Alexandru Ciolan, Pedro García-Sánchez and the second author. In an appendix Pedro García-Sánchez shows that for every $k\ge 4$ there exists a symmetric non-cyclotomic numerical semigroup having Frobenius number $2k+1.$

math.NT↗

arXiv2017

Genetic and Memetic Algorithm with Diversity Equilibrium based on Greedy Diversification

The lack of diversity in a genetic algorithm's population may lead to a bad performance of the genetic operators since there is not an equilibrium between exploration and exploitation. In those cases, genetic algorithms present a fast and unsuitable convergence. In this paper we develop a novel hybrid genetic algorithm which attempts to obtain a balance between exploration and exploitation. It confronts the diversity problem using the named greedy diversification operator. Furthermore, the proposed algorithm applies a competition between parent and children so as to exploit the high quality visited solutions. These operators are complemented by a simple selection mechanism designed to preserve and take advantage of the population diversity. Additionally, we extend our proposal to the field of memetic algorithms, obtaining an improved model with outstanding results in practice. The experimental study shows the validity of the approach as well as how important is taking into account the exploration and exploitation concepts when designing an evolutionary algorithm.

cs.AI↗