Artificial Intelligence最新文献

{"title":"Human compliance with computational argumentation principles","authors":"Predrag Teovanović ,&nbsp;Srdjan Vesic ,&nbsp;Bruno Yun","doi":"10.1016/j.artint.2025.104457","DOIUrl":"10.1016/j.artint.2025.104457","url":null,"abstract":"<div><div>This paper presents a comprehensive examination of human compliance with normative principles of argumentation across two experimental studies. The first study investigated whether fundamental argumentation principles such as anonymity, independence, void precedence, and maximality align with human reasoning. Additionally, it explored whether graph-based representations of arguments facilitate better understanding and adherence to these principles compared to textual representations of arguments alone and examined the role of individual cognitive differences in compliance with these principles. Our experiments revealed that graph-based representations significantly improved compliance with argumentation principles, particularly among individuals with higher cognitive reflection. The second study replicated and extended the first study’s findings, introducing new principles such as skeptical precedence and simple reinstatement, and explored the effects of presenting arguments solely in graphical form, as well as the impact of a short tutorial on argumentation theory. The study also assessed participants’ ability to perform graphical tasks and how this influenced their compliance with normative principles. Results partially replicated the first study’s findings, confirming that graphical representations enhance compliance, but also revealed that the effect does not generalize to the new principles. We found evidence that in the absence of a graphical representation, performing graphical tasks can improve compliance with principles; especially drawing the argumentation graph. Moreover, a brief tutorial significantly improved performance on several principles, indicating that even minimal instruction can enhance understanding and compliance. However, the difficulties observed with the simple reinstatement principle hint that the participants’ intuition about the notion of defense diverges significantly from that of the researchers and that more careful thoughts must be put in crafting them. These studies collectively suggest that while argumentation principles can be intuitive to some extent, their comprehension and application are significantly influenced by the instruction given as well as by graphical representations and processes used to obtain them. These findings have important implications for the design of future argumentation-based tools and our understanding of how to bridge human reasoning and formal argumentation.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"351 ","pages":"Article 104457"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145560017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-objective reinforcement learning for provably incentivising alignment with value systems","authors":"Manel Rodriguez-Soto ,&nbsp;Roxana Rădulescu ,&nbsp;Filippo Bistaffa ,&nbsp;Oriol Ricart ,&nbsp;Arnau Mayoral-Macau ,&nbsp;Maite Lopez-Sanchez ,&nbsp;Juan A. Rodriguez-Aguilar ,&nbsp;Ann Nowé","doi":"10.1016/j.artint.2025.104460","DOIUrl":"10.1016/j.artint.2025.104460","url":null,"abstract":"<div><div>This paper addresses the problem of ensuring that autonomous learning agents align with multiple moral values. Specifically, we present the theoretical principles and algorithmic tools necessary for creating an environment where we ensure that the agent learns a behaviour aligned with multiple moral values while striving to achieve its individual objective. To address this value alignment problem, we adopt the Multi-Objective Reinforcement Learning framework and propose a novel algorithm that combines techniques from Multi-Objective Reinforcement Learning and Linear Programming. In addition, we illustrate our value alignment process with an example involving an autonomous vehicle. Here, we demonstrate that the agent learns to behave in alignment with the ethical values of safety, achievement, and comfort, with achievement representing the agent’s individual objective. Such ethical behaviour differs depending on the ordering between values. We also use a synthetic multi-objective environment to evaluate the computational costs of guaranteeing ethical learning as the number of values increases.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"351 ","pages":"Article 104460"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145689753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Federated neural nonparametric point processes","authors":"Hui Chen ,&nbsp;Xuhui Fan ,&nbsp;Hengyu Liu ,&nbsp;Yaqiong Li ,&nbsp;Zhilin Zhao ,&nbsp;Feng Zhou ,&nbsp;Christopher John Quinn ,&nbsp;Longbing Cao","doi":"10.1016/j.artint.2025.104454","DOIUrl":"10.1016/j.artint.2025.104454","url":null,"abstract":"<div><div>Temporal point processes (TPPs) are effective for modeling event occurrences over time but struggle with sparse and uncertain events in federated systems, where privacy is a major concern. To address this, we propose <em>FedPP</em>, a federated neural nonparametric point process model. FedPP integrates neural embeddings into sigmoidal Gaussian Cox processes (SGCPs) on the client side. SGCPs is a flexible and expressive class of TPPs, allowing FedPP to generate highly flexible intensity functions that capture client-specific event dynamics and uncertainties while efficiently summarizing historical records. For global aggregation, FedPP introduces a divergence-based mechanism to communicate the distributions of kernel hyperparameters in SGCPs between the server and clients, while keeping client-specific parameters local to ensure privacy and personalization. FedPP effectively captures event uncertainty and sparsity. Extensive experiments demonstrate its superior performance in federated settings, showing global aggregation with the KL divergence and the Wasserstein distance.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"351 ","pages":"Article 104454"},"PeriodicalIF":4.6,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145575241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bridging sparse domain semantics via an asymmetric siamese framework with virtual anchor guidance for domain-specific multimodal translation","authors":"Junjun Guo ,&nbsp;Yifan Liu ,&nbsp;Zhengtao Yu","doi":"10.1016/j.artint.2025.104443","DOIUrl":"10.1016/j.artint.2025.104443","url":null,"abstract":"<div><div>Domain-specific Multimodal Neural Machine Translation (DMNMT) aims to translate text in specialized domains by leveraging both linguistic context and associated visual information to resolve domain-specific ambiguities and enhance terminological accuracy. Although accompanying images often provide sparse and fragmented visual cues that could potentially anchor critical domain semantics, the semantic mapping from images to textual domain semantics typically exhibits sparse multi-focal alignment challenges. Existing general-domain multimodal neural machine translation (MNMT) models and large language models (LLMs) struggle to achieve accurate aggregation of domain-salient information, often resulting in near-equivalent yet imprecise terminology translations or outright errors. To bridge this sparse domain semantic correspondence gap, we introduce the Asymmetric Siamese Multimodal Fusion (ASMF) framework, which decouples domain representation learning into two complementary branches that both consume text: a domain-specific virtual visual content generation (DVVG) branch and a terminology-aware textual (TAT) branch. The DVVG branch distills sparse, localized visual features into modality-agnostic semantic anchors through mask-constrained multi-focal distillation, while the TAT branch captures terminology-dense textual context. We introduce a novel Domain-Virtualized Pivot-driven Hierarchical Fusion (DVPH) strategy that progressively injects distilled visual anchors across encoder layers. This asymmetric dual-branch design effectively couples spatially fragmented visual details with terminology-rich text, enabling accurate and domain-consistent translations even for low-frequency terms. Extensive experiments were conducted on four benchmark datasets covering three distinct scenarios: two domain-specific datasets (Fashion-MMT and EMMT), one general-domain dataset (Multi30K), and one multi-domain dataset (WIT). Comprehensive evaluations demonstrate that the proposed approach outperforms existing MNMT, DMNMT and LLMs, achieving state-of-the-art (SOTA) results across all datasets. In-depth analyses validate its robustness and generalization capabilities across diverse scenarios, including visually noisy or image-free conditions.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"350 ","pages":"Article 104443"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145404964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Many-objective problems where crossover is provably essential","authors":"Andre Opris","doi":"10.1016/j.artint.2025.104453","DOIUrl":"10.1016/j.artint.2025.104453","url":null,"abstract":"<div><div>This article addresses theory in evolutionary many-objective optimization and focuses on the role of crossover operators. The advantages of using crossover are hardly understood and rigorous runtime analyses with crossover are lagging far behind its use in practice, specifically in the case of more than two objectives. We present two many-objective problems <span><math><msub><mtext>RR</mtext><mrow><mi>MO</mi></mrow></msub></math></span> and <span><math><msub><mtext>URR</mtext><mrow><mi>MO</mi></mrow></msub></math></span>, and a theoretical runtime analysis of the GSEMO and the widely used NSGA‑III algorithm, to demonstrate that one point crossover on <span><math><msub><mtext>RR</mtext><mrow><mi>MO</mi></mrow></msub></math></span>, as well as uniform crossover on <span><math><msub><mtext>URR</mtext><mrow><mi>MO</mi></mrow></msub></math></span>, can yield an exponential speedup in the runtime. In particular, when the number of objectives is constant, this algorithms can find the Pareto set of both problems in expected polynomial time when using crossover, while without crossover they require exponential time to even find a single Pareto-optimal point. For either problem, we also demonstrate a significant performance gap in certain superconstant parameter regimes for the number of objectives. To the best of our knowledge, this is the first rigorous runtime analysis in many-objective optimization which demonstrates an exponential performance gap when using crossover for more than two objectives. Additionally, it is the first runtime analysis involving crossover in many-objective optimization where the number of objectives is not necessarily constant.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"350 ","pages":"Article 104453"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145461589","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A General Theoretical Framework for Learning Smallest Interpretable Models","authors":"Sebastian Ordyniak ,&nbsp;Giacomo Paesani ,&nbsp;Mateusz Rychlicki ,&nbsp;Stefan Szeider","doi":"10.1016/j.artint.2025.104441","DOIUrl":"10.1016/j.artint.2025.104441","url":null,"abstract":"<div><div>We develop a general algorithmic framework that allows us to obtain fixed-parameter tractability for computing smallest symbolic models that represent given data. Our framework applies to all ML model types that admit a certain extension property. By establishing this extension property for decision trees, decision sets, decision lists, and binary decision diagrams, we obtain that minimizing these fundamental model types is fixed-parameter tractable. Our framework even applies to ensembles, which combine individual models by majority decision.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"350 ","pages":"Article 104441"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145382554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Kernel-bounded clustering: Achieving the objective of spectral clustering without eigendecomposition","authors":"Hang Zhang ,&nbsp;Kai Ming Ting ,&nbsp;Ye Zhu","doi":"10.1016/j.artint.2025.104440","DOIUrl":"10.1016/j.artint.2025.104440","url":null,"abstract":"<div><div>The research on spectral clustering (SC) has thus far been pursued on the same track using the same tool of eigendecomposition of a matrix since the idea was first introduced in 1973. Despite its successes, SC has been identified to have fundamental limitations that prevent SC from discovering certain types of clusters, and SC has slow runtime. We offer an alternative path that does not involve the eigendecomposition, and, more broadly, it uses no optimization. The proposed new Kernel-Bounded Clustering (KBC) is a complete metamorphosis in 50 years of research in SC in view of the fact that KBC achieves the same objective of SC without eigendecomposition or optimization. We evaluated KBC on the datasets that have been used to demonstrate the fundamental limitations of SC, genome-wide expression data, large image datasets and many commonly used real-world benchmark datasets. KBC produced better quality clusters than various variants of SC, and it ran six orders of magnitude faster than the traditional SC on a set of 5 million data points.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"350 ","pages":"Article 104440"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145361111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Online POMDP planning with anytime deterministic optimality guarantees","authors":"Moran Barenboim ,&nbsp;Vadim Indelman","doi":"10.1016/j.artint.2025.104442","DOIUrl":"10.1016/j.artint.2025.104442","url":null,"abstract":"<div><div>Decision-making under uncertainty is a critical aspect of many practical autonomous systems due to incomplete information. Partially Observable Markov Decision Processes (POMDPs) offer a mathematically principled framework for formulating decision-making problems under such conditions. However, finding an optimal solution for a POMDP is generally intractable. In recent years, there has been a significant progress of scaling approximate solvers from small to moderately sized problems, using online tree search solvers. Often, such approximate solvers are limited to probabilistic or asymptotic guarantees towards the optimal solution. In this paper, we derive a deterministic relationship for discrete POMDPs between an approximated and the optimal solution. We show that at any time, we can derive bounds that relate between the existing solution and the optimal one. We show that our derivations provide an avenue for a new set of algorithms and can be attached to existing algorithms that have a certain structure to provide them with deterministic guarantees with marginal computational overhead. In return, not only do we certify the solution quality, but we demonstrate that making a decision based on the deterministic guarantee may result in superior performance compared to the original algorithm without the deterministic certification.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"350 ","pages":"Article 104442"},"PeriodicalIF":4.6,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145382555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Defending a city from multi-drone attacks: A sequential Stackelberg security games approach","authors":"Dolev Mutzari ,&nbsp;Tonmoay Deb ,&nbsp;Cristian Molinaro ,&nbsp;Andrea Pugliese ,&nbsp;V.S. Subrahmanian ,&nbsp;Sarit Kraus","doi":"10.1016/j.artint.2025.104425","DOIUrl":"10.1016/j.artint.2025.104425","url":null,"abstract":"<div><div>To counter an imminent multi-drone attack on a city, defenders have deployed drones across the city. These drones must intercept/eliminate the threat, thus reducing potential damage from the attack. We model this as a Sequential Stackelberg Security Game, where the defender first commits to a mixed sequential defense strategy, and the attacker then best responds. We develop an efficient algorithm called S2D2, which outputs a defense strategy. We demonstrate the efficacy of S2D2 in extensive experiments on data from 80 real cities, improving the performance of the defender in comparison to greedy heuristics based on prior works. We prove that under some reasonable assumptions about the city structure, S2D2 outputs an approximate Strong Stackelberg Equilibrium (SSE) with a convenient structure.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"349 ","pages":"Article 104425"},"PeriodicalIF":4.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145263925","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Contra2: A one-step active learning method for imbalanced graphs","authors":"Wenjie Yang ,&nbsp;Shengzhong Zhang ,&nbsp;Jiaxing Guo ,&nbsp;Zengfeng Huang","doi":"10.1016/j.artint.2025.104439","DOIUrl":"10.1016/j.artint.2025.104439","url":null,"abstract":"<div><div>Graph active learning (GAL) is an important research direction in graph neural networks (GNNs) that aims to select the most valuable nodes for labeling to train GNNs. Previous works in GAL have primarily focused on the overall performance of GNNs, overlooking the balance among different classes. However, graphs in real-world applications are often imbalanced, which leads GAL methods to select class-imbalanced training sets, resulting in biased GNN models. Furthermore, due to the high cost of multi-turn queries, there is an increasing demand for one-step GAL methods, where the entire training set is queried at once. These realities prompt us to investigate the problem of one-step active learning on imbalanced graphs.</div><div>In this paper, we propose a theory-driven method called Contrast &amp; Contract (Contra²) to tackle the above issues. The key idea of Contra² is that intra-class edges within the majority are dominant in the edge set, so contracting these edges will reduce the imbalance ratio. Specifically, Contra² first learns node representations by graph <strong>contrast</strong>ive learning (GCL), then stochastically <strong>contract</strong>s the edges that connect nodes with similar embeddings. We theoretically show that Contra² reduces the imbalance ratio with high probability. By leveraging a more evenly distributed graph, we can achieve a balanced selection of labeled nodes without requiring any seed labels. The effectiveness of Contra² is evaluated against various baselines on 11 datasets with different budgets. Contra² demonstrates remarkable performance, achieving either higher or on-par performance with only half of the annotation budget on some datasets.</div></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"349 ","pages":"Article 104439"},"PeriodicalIF":4.6,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145263924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}