arXiv subjects

Miguel Couceiro

Publications and source records attributed to Miguel Couceiro.

At least 19 recordsLinked to original sources

arXiv2026

EntmaxKV: Support-Aware Decoding for Entmax Attention

Long-context decoding is increasingly limited by KV-cache memory traffic since each generated token attends over a cache whose size grows linearly with context length. Existing sparse decoding methods reduce this cost by selecting subsets of tokens or pages, but are designed for softmax attention, whose dense tails make any truncation discard nonzero probability mass. In contrast, $α$-entmax produces exact zeros, turning sparse decoding from dense-tail approximation into support recovery: if the selected candidates contain the entmax support, sparse decoding remains exact. While recent entmax kernels enable efficient training, they do not address the autoregressive decoding bottleneck, where dense inference still streams the full KV cache before sparsity is known. In this work, we introduce EntmaxKV, an entmax-native sparse decoding framework that exploits sparsity before KV pages are loaded. EntmaxKV combines query-aware page scoring, support-aware candidate selection, and sparse entmax attention. We analyze truncation error through the dropped probability mass $δ$, showing that output error is controlled by $δ$ and vanishes when the entmax support is recovered. We further introduce a Gaussian-aware entmax selector that estimates the entmax threshold from lightweight page statistics, adapting the selected budget to the score distribution. Empirically, EntmaxKV drops less probability mass, retains more support tokens, and achieves lower output error than softmax-based sparse decoding at matched KV budgets. On long-context and language modeling benchmarks, it closely matches full-cache entmax while using a small fraction of the KV cache, achieving up to $3.36\times$ (softmax) and $5.43\times$ (entmax) speedup over full attention baselines at 1M context length. Code available at: https://github.com/deep-spin/entmaxkv.

cs.LG↗

arXiv2026

Shapley Regression for Rare Disease Diagnosis Support: a case study on APDS

Activated PI3K8 Syndrome (APDS) is a rare genetic immune disorder caused by variants in PIK3CD or PIK3R1, with highly heterogeneous symptoms that often delay diagnosis. Early recognition is hampered by overlapping clinical presentations and limited clinician awareness, motivating systematic, data-driven approaches to detect APDS-associated phenotypic patterns in routine electronic health records. Traditional linear scoring systems cannot capture complex symptom interactions, while deep learning models, though expressive, often lack interpretability. To bridge this gap, we propose Shapley regression, a novel game-theoretic model replacing the linear predictor with a k-additive cooperative game, explicitly modeling co-occurrence of symptoms while maintaining the transparency and convexity of logistic regression. We carry out an empirical study of our lightweight method on eight public biomedical datasets, showing that a 2-additive model with $l_{2}$ regularization achieves an optimal trade-off between predictive power and noise robustness. We also apply it to a real-world cohort of 222 patients, on which Shapley regression accurately distinguished APDS cases from matched controls, confirming and validating phenotypes known to be associated with APDS, and facilitating the exploration of pairwise interactions between symptoms, validated by clinical experts.

cs.LG↗

arXiv2026

Which Are the Low-Resource Languages of the Semantic Web?

Emerging digital technologies are exacerbating the existing divide in Open Access Data (OAD) between high-and low-resource languages, excluding many communities from the global digital transformation. Multilingual Linked Open Data Knowledge Graphs (LOD KGs) could contribute to mitigating this divide through cross-lingual transfer; however, no clear quantitative definition of low-resource languages has yet been established in the context of LOD KGs. In this poster, we present a methodology to analyze the distribution of languages across LOD KGs and propose a preliminary multi-level categorization based on DBpedia, BabelNet, and Wikidata. This categorization is leveraged to bring a formal definition of low-, high-, and medium-resource languages that could be later leveraged to select cross-lingual transfer candidates.

cs.AI↗

arXiv2026

FrameNet Semantic Role Classification by Analogy

In this paper, we adopt a relational view of analogies applied to Semantic Role Classification in FrameNet. We define analogies as formal relations over the Cartesian product of frame evoking lexical units (LUs) and frame element (FEs) pairs, which we use to construct a new dataset. Each element of this binary relation is labelled as a valid analogical instance if the frame elements share the same semantic role, or as invalid otherwise. This formulation allows us to transform Semantic Role Classification into binary classification and train a lightweight Artificial Neural Network (ANN) that exhibits rapid convergence with minimal parameters. Unconventionally, no Semantic Role information is introduced to the neural network during training. We recover semantic roles during inference by computing probability distributions over candidates of all semantic roles within a given frame through random sampling and analogical transfer. This approach allows us to surpass previous state-of-the-art results while maintaining computational efficiency and frugality.

cs.CL↗

arXiv2026

A hybrid approach for building fuzzy numbers based on data and expert knowledge

This paper presents a hybrid socio-technical methodology for constructing fuzzy numbers from numerical data while incorporating expert knowledge through an interactive Deck of Cards (DoC) process. The approach extends the existing DoC membership function construction framework by introducing a data-driven pipeline based on a convex version of fuzzy $k$-Means in which each computational step produces intermediate outputs that are translated into card-based structures for expert validation and tuning. The proposed method ensures interpretability, adaptability, and consistency between empirical evidence and expert semantics.

math.GM↗

arXiv2025

New Perspectives on Semiring Applications to Dynamic Programming

Semiring algebras have been shown to provide a suitable language to formalize many noteworthy combinatorial problems. For instance, the Shortest-Path problem can be seen as a special case of the Algebraic-Path problem when applied to the tropical semiring. The application of semirings typically makes it possible to solve extended problems without increasing the computational complexity. In this article we further exploit the idea of using semiring algebras to address and tackle several extensions of classical computational problems by dynamic programming. We consider a general approach which allows us to define a semiring extension of any problem with a reasonable notion of a certificate (e.g., an NP problem). This allows us to consider cost variants of these combinatorial problems, as well as their counting extensions where the goal is to determine how many solutions a given problem admits. The approach makes no particular assumptions (such as idempotence) on the semiring structure. We also propose a new associative algebraic operation on semirings, called $Δ$-product, which enables our dynamic programming algorithms to count the number of solutions of minimal costs. We illustrate the advantages of our framework on two well-known but computationally very different NP-hard problems, namely, Connected-Dominating-Set problems and finite-domain Constraint Satisfaction Problems (CSPs). In particular, we prove fixed parameter tractability (FPT) with respect to clique-width and tree-width of the input. This also allows us to count solutions of minimal cost, which is an overlooked problem in the literature.

cs.CC↗

arXiv2025

Generalizing Analogical Inference from Boolean to Continuous Domains

Analogical reasoning is a powerful inductive mechanism, widely used in human cognition and increasingly applied in artificial intelligence. Formal frameworks for analogical inference have been developed for Boolean domains, where inference is provably sound for affine functions and approximately correct for functions close to affine. These results have informed the design of analogy-based classifiers. However, they do not extend to regression tasks or continuous domains. In this paper, we revisit analogical inference from a foundational perspective. We first present a counterexample showing that existing generalization bounds fail even in the Boolean setting. We then introduce a unified framework for analogical reasoning in real-valued domains based on parameterized analogies defined via generalized means. This model subsumes both Boolean classification and regression, and supports analogical inference over continuous functions. We characterize the class of analogy-preserving functions in this setting and derive both worst-case and average-case error bounds under smoothness assumptions. Our results offer a general theory of analogical inference across discrete and continuous domains.

cs.AI↗

arXiv2025

Improved Bounds for Twin-Width Parameter Variants with Algorithmic Applications to Counting Graph Colorings

The $H$-Coloring problem is a well-known generalization of the classical NP-complete problem $k$-Coloring where the task is to determine whether an input graph admits a homomorphism to the template graph $H$. This problem has been the subject of intense theoretical research and in this article we study the complexity of $H$-Coloring with respect to the parameters clique-width and the more recent component twin-width, which describe desirable computational properties of graphs. We give two surprising linear bounds between these parameters, thus improving the previously known exponential and double exponential bounds. Our constructive proof naturally extends to related parameters and as a showcase we prove that total twin-width and linear clique-width can be related via a tight quadratic bound. These bounds naturally lead to algorithmic applications. The linear bounds between component twin-width and clique-width entail natural approximations of component twin-width, by making use of the results known for clique-width. As for computational aspects of graph coloring, we target the richer problem of counting the number of homomorphisms to $H$ (#$H$-Coloring). The first algorithm that we propose uses a contraction sequence of the input graph $G$ parameterized by the component twin-width of $G$. This leads to a positive FPT result for the counting version. The second uses a contraction sequence of the template graph $H$ and here we instead measure the complexity with respect to the number of vertices in the input graph. Using our linear bounds we show that our algorithms are always at least as fast as the previously best #$H$-Coloring algorithms (based on clique-width) and for several interesting classes of graphs (e.g., cographs, cycles of length $\ge 7$, or distance-hereditary graphs) are in fact strictly faster.

cs.CC↗

arXiv2025

Comparing representations of long clinical texts for the task of patient note-identification

In this paper, we address the challenge of patient-note identification, which involves accurately matching an anonymized clinical note to its corresponding patient, represented by a set of related notes. This task has broad applications, including duplicate records detection and patient similarity analysis, which require robust patient-level representations. We explore various embedding methods, including Hierarchical Attention Networks (HAN), three-level Hierarchical Transformer Networks (HTN), LongFormer, and advanced BERT-based models, focusing on their ability to process mediumto-long clinical texts effectively. Additionally, we evaluate different pooling strategies (mean, max, and mean_max) for aggregating wordlevel embeddings into patient-level representations and we examine the impact of sliding windows on model performance. Our results indicate that BERT-based embeddings outperform traditional and hierarchical models, particularly in processing lengthy clinical notes and capturing nuanced patient representations. Among the pooling strategies, mean_max pooling consistently yields the best results, highlighting its ability to capture critical features from clinical notes. Furthermore, the reproduction of our results on both MIMIC dataset and Necker hospital data warehouse illustrates the generalizability of these approaches to real-world applications, emphasizing the importance of both embedding methods and aggregation strategies in optimizing patient-note identification and enhancing patient-level modeling.

cs.CL↗

arXiv2025

Exploring ReAct Prompting for Task-Oriented Dialogue: Insights and Shortcomings

Large language models (LLMs) gained immense popularity due to their impressive capabilities in unstructured conversations. Empowering LLMs with advanced prompting strategies such as reasoning and acting (ReAct) (Yao et al., 2022) has shown promise in solving complex tasks traditionally requiring reinforcement learning. In this work, we apply the ReAct strategy to guide LLMs performing task-oriented dialogue (TOD). We evaluate ReAct-based LLMs (ReAct-LLMs) both in simulation and with real users. While ReAct-LLMs severely underperform state-of-the-art approaches on success rate in simulation, this difference becomes less pronounced in human evaluation. Moreover, compared to the baseline, humans report higher subjective satisfaction with ReAct-LLM despite its lower success rate, most likely thanks to its natural and confidently phrased responses.

cs.CL↗

arXiv2025

Unveiling Biases while Embracing Sustainability: Assessing the Dual Challenges of Automatic Speech Recognition Systems

In this paper, we present a bias and sustainability focused investigation of Automatic Speech Recognition (ASR) systems, namely Whisper and Massively Multilingual Speech (MMS), which have achieved state-of-the-art (SOTA) performances. Despite their improved performance in controlled settings, there remains a critical gap in understanding their efficacy and equity in real-world scenarios. We analyze ASR biases w.r.t. gender, accent, and age group, as well as their effect on downstream tasks. In addition, we examine the environmental impact of ASR systems, scrutinizing the use of large acoustic models on carbon emission and energy consumption. We also provide insights into our empirical analyses, offering a valuable contribution to the claims surrounding bias and sustainability in ASR systems.

cs.CL↗

arXiv2025

Resistive Plate Chambers for brain PET imaging and particle tracking and timing (TOF-tracker)

In this work we explore readout architectures for the simultaneous high-resolution timing and bidimensional tracking of charged particles with Resistive Plate Chambers (TOF-tracker) and for the accurate detection of gamma rays for PET imaging. On 625 $cm^2$ of active area we obtained a time resolution of 61 ps $σ$ and bidimensional position resolution below 150 $μ$m $σ$ for the tracking and timing of charged particles from cosmic rays. An intrinsic precision of 0.49 mm FWHM was determined for the localization of a small $β^+$ source via the detection of its annihilation radiation.

physics.ins-det↗

arXiv2024

KGPrune: a Web Application to Extract Subgraphs of Interest from Wikidata with Analogical Pruning

Knowledge graphs (KGs) have become ubiquitous publicly available knowledge sources, and are nowadays covering an ever increasing array of domains. However, not all knowledge represented is useful or pertaining when considering a new application or specific task. Also, due to their increasing size, handling large KGs in their entirety entails scalability issues. These two aspects asks for efficient methods to extract subgraphs of interest from existing KGs. To this aim, we introduce KGPrune, a Web Application that, given seed entities of interest and properties to traverse, extracts their neighboring subgraphs from Wikidata. To avoid topical drift, KGPrune relies on a frugal pruning algorithm based on analogical reasoning to only keep relevant neighbors while pruning irrelevant ones. The interest of KGPrune is illustrated by two concrete applications, namely, bootstrapping an enterprise KG and extracting knowledge related to looted artworks.

cs.AI↗

arXiv2024

REFINE-LM: Mitigating Language Model Stereotypes via Reinforcement Learning

With the introduction of (large) language models, there has been significant concern about the unintended bias such models may inherit from their training data. A number of studies have shown that such models propagate gender stereotypes, as well as geographical and racial bias, among other biases. While existing works tackle this issue by preprocessing data and debiasing embeddings, the proposed methods require a lot of computational resources and annotation effort while being limited to certain types of biases. To address these issues, we introduce REFINE-LM, a debiasing method that uses reinforcement learning to handle different types of biases without any fine-tuning. By training a simple model on top of the word probability distribution of a LM, our bias agnostic reinforcement learning method enables model debiasing without human annotations or significant computational resources. Experiments conducted on a wide range of models, including several LMs, show that our method (i) significantly reduces stereotypical biases while preserving LMs performance; (ii) is applicable to different types of biases, generalizing across contexts such as gender, ethnicity, religion, and nationality-based biases; and (iii) it is not expensive to train.

cs.CL↗

arXiv2024

Any four real numbers are on all fours with analogy

This work presents a formalization of analogy on numbers that relies on generalized means. It is motivated by recent advances in artificial intelligence and applications of machine learning, where the notion of analogy is used to infer results, create data and even as an assessment tool of object representations, or embeddings, that are basically collections of numbers (vectors, matrices, tensors). This extended analogy use asks for mathematical foundations and clear understanding of the notion of analogy between numbers. We propose a unifying view of analogies that relies on generalized means defined in terms of a power parameter. In particular, we show that any four increasing positive real numbers is an analogy in a unique suitable power. In addition, we show that any such analogy can be reduced to an equivalent arithmetic analogy and that any analogical equation has a solution for increasing numbers, which generalizes without restriction to complex numbers. These foundational results provide a better understanding of analogies in areas where representations are numerical.

cs.AI↗

arXiv2024

Uncertainty Management in the Construction of Knowledge Graphs: a Survey

Knowledge Graphs (KGs) are a major asset for companies thanks to their great flexibility in data representation and their numerous applications, e.g., vocabulary sharing, Q/A or recommendation systems. To build a KG it is a common practice to rely on automatic methods for extracting knowledge from various heterogeneous sources. But in a noisy and uncertain world, knowledge may not be reliable and conflicts between data sources may occur. Integrating unreliable data would directly impact the use of the KG, therefore such conflicts must be resolved. This could be done manually by selecting the best data to integrate. This first approach is highly accurate, but costly and time-consuming. That is why recent efforts focus on automatic approaches, which represents a challenging task since it requires handling the uncertainty of extracted knowledge throughout its integration into the KG. We survey state-of-the-art approaches in this direction and present constructions of both open and enterprise KGs and how their quality is maintained. We then describe different knowledge extraction methods, introducing additional uncertainty. We also discuss downstream tasks after knowledge acquisition, including KG completion using embedding models, knowledge alignment, and knowledge fusion in order to address the problem of knowledge uncertainty in KG construction. We conclude with a discussion on the remaining challenges and perspectives when constructing a KG taking into account uncertainty.

cs.AI↗

arXiv2024

On the Calibration of Epistemic Uncertainty: Principles, Paradoxes and Conflictual Loss

The calibration of predictive distributions has been widely studied in deep learning, but the same cannot be said about the more specific epistemic uncertainty as produced by Deep Ensembles, Bayesian Deep Networks, or Evidential Deep Networks. Although measurable, this form of uncertainty is difficult to calibrate on an objective basis as it depends on the prior for which a variety of choices exist. Nevertheless, epistemic uncertainty must in all cases satisfy two formal requirements: first, it must decrease when the training dataset gets larger and, second, it must increase when the model expressiveness grows. Despite these expectations, our experimental study shows that on several reference datasets and models, measures of epistemic uncertainty violate these requirements, sometimes presenting trends completely opposite to those expected. These paradoxes between expectation and reality raise the question of the true utility of epistemic uncertainty as estimated by these models. A formal argument suggests that this disagreement is due to a poor approximation of the posterior distribution rather than to a flaw in the measure itself. Based on this observation, we propose a regularization function for deep ensembles, called conflictual loss in line with the above requirements. We emphasize its strengths by showing experimentally that it restores both requirements of epistemic uncertainty, without sacrificing either the performance or the calibration of the deep ensembles.

cs.LG↗

arXiv2024

Solving morphological analogies: from retrieval to generation

Analogical inference is a remarkable capability of human reasoning, and has been used to solve hard reasoning tasks. Analogy based reasoning (AR) has gained increasing interest from the artificial intelligence community and has shown its potential in multiple machine learning tasks such as classification, decision making and recommendation with competitive results. We propose a deep learning (DL) framework to address and tackle two key tasks in AR: analogy detection and solving. The framework is thoroughly tested on the Siganalogies dataset of morphological analogical proportions (APs) between words, and shown to outperform symbolic approaches in many languages. Previous work have explored the behavior of the Analogy Neural Network for classification (ANNc) on analogy detection and of the Analogy Neural Network for retrieval (ANNr) on analogy solving by retrieval, as well as the potential of an autoencoder (AE) for analogy solving by generating the solution word. In this article we summarize these findings and we extend them by combining ANNr and the AE embedding model, and checking the performance of ANNc as an retrieval method. The combination of ANNr and AE outperforms the other approaches in almost all cases, and ANNc as a retrieval method achieves competitive or better performance than 3CosMul. We conclude with general guidelines on using our framework to tackle APs with DL.

cs.CL↗