Hasil untuk "Semantics"

K. Hale, M. Kenstowicz

M. Maloney, E. A. Lanier, Robert K. Straus

Ayan Sar, Tanupriya Choudhury, Sampurna Roy et al.

Few-shot multispectral object detection remains a formidable challenge in remote sensing, constrained by the scarcity of annotated data across heterogeneous modalities and environmental conditions. Existing transformer-based detectors, while powerful, often exhibit overfitting in low-sample regimes and fail to preserve cross-spectral consistency between visible and infrared channels. To address these limitations, this article presents few-shot spatial–spectral prototype calibration network (Few-SSPC-Net), a spatial–spectral prototype calibration network designed for efficient and adaptive few-shot multispectral object detection. Unlike conventional transformer-driven pipelines, this framework employs a transformer-free dual-branch convolutional architecture—one branch emphasizing spatial semantics and the other spectral correlations—bridged by a Cross-Scale Interaction Module for fine-grained feature alignment across modalities. Central to this framework is the proposed Spatial–Spectral Prototype Calibration Module, which dynamically refines class prototypes through spectral correlation-guided calibration between support and query samples. This mechanism mitigates prototype drift and enhances generalization by enforcing spectral angular consistency within the embedding space. The entire architecture is trained under an episodic meta-learning paradigm, optimizing a joint objective of classification, localization, and spectral calibration regularization. Extensive experiments on benchmark datasets demonstrate that Few-SSPC-Net achieves consistent gains over state-of-the-art few-shot detectors, with up to +4.7% mAP improvement under five-shot settings, while maintaining competitive inference efficiency. The results affirm the positioning of Few-SSPC-Net as a robust framework for multispectral object detection in complex, data-limited remote sensing scenarios.

Petr Biskup

This article concerns the morpheme <i>a</i> in Czech. It occurs in nominals, conjunctions, and various verbal predicates. In contrast to the common practice of treating such <i>a</i> exponents as independent, accidentally homophonous elements, it is argued that some of these <i>a</i>s can be treated as one item. What the syncretic <i>a</i>s have in common is pluralizing semantics. Thus, the article proposes that verbal number (specifically, plurality) is related to nominal number and conjunctions. The article addresses the questions of how the multifunctionality of morphemes—such as the Czech <i>a</i>—can be analyzed and which tools of lexical–realizational approaches to morphology are most suitable for the analysis. In addition to the plural interpretation, <i>a</i> brings about changes in the argument structure of verbal predicates and fulfills several functions in the nominal and conjunction domains. The analysis is couched in the Distributed Morphology framework. However, contrary to expectations, the multifunctional <i>a</i> is not treated as an underspecified marker. It is analyzed as an overspecified marker that can realize (i.e., span) several syntactic heads: the pluralization head with the pluralization operator, the voice head, plus some other heads present in verbs and nominals. It is argued that the best option for deriving the multifunctional property of <i>a</i> is to assume the superset principle and pre-linearization spanning.

Julien Saan Joachim, Marc de Visme, Stefan Haar et al.

Classical Petri nets provide a canonical model of concurrency, with unfolding semantics linking nets, occurrence nets, and event structures. No comparable framework exists for quantum concurrency: existing ''quantum Petri nets'' lack rigorous concurrent and sound quantum semantics, analysis tools, and unfolding theory. We introduce Quantum Petri Nets (QPNs), Petri nets equipped with a quantum valuation compatible with the quantum event structure semantics of Clairambault, De Visme, and Winskel (2019). Our contributions are: (i) a local definition of Quantum Occurrence Nets (LQONs) compatible with quantum event structures, (ii) a construction of QPNs with a well-defined unfolding semantics, (iii) a compositional framework for QPNs. This establishes a semantically well grounded model of quantum concurrency, bridging Petri net theory and quantum programming.

Yueming Pan, Ruoyu Feng, Qi Dai et al.

Latent Diffusion Models (LDMs) inherently follow a coarse-to-fine generation process, where high-level semantic structure is generated slightly earlier than fine-grained texture. This indicates the preceding semantics potentially benefit texture generation by providing a semantic anchor. Recent advances have integrated semantic priors from pretrained visual encoders to further enhance LDMs, yet they still denoise semantic and VAE-encoded texture synchronously, neglecting such ordering. Observing these, we propose Semantic-First Diffusion (SFD), a latent diffusion paradigm that explicitly prioritizes semantic formation. SFD first constructs composite latents by combining a compact semantic latent, which is extracted from a pretrained visual encoder via a dedicated Semantic VAE, with the texture latent. The core of SFD is to denoise the semantic and texture latents asynchronously using separate noise schedules: semantics precede textures by a temporal offset, providing clearer high-level guidance for texture refinement and enabling natural coarse-to-fine generation. On ImageNet 256x256 with guidance, SFD achieves FID 1.06 (LightningDiT-XL) and FID 1.04 (1.0B LightningDiT-XXL), while achieving up to 100x faster convergence than the original DiT. SFD also improves existing methods like ReDi and VA-VAE, demonstrating the effectiveness of asynchronous, semantics-led modeling. Project page and code: https://yuemingpan.github.io/SFD.github.io/.

Laczházi Gyula

The distinction between the necessary and the accidental, between events that serve a purpose and those that are meaningless, can be found in the semantics developed by a wide variety of cultures to interpret the world. In the religiously dominated culture of the Middle Ages and the early modern period, the question of contingency was thematized primarily in relation to the concept of providence, and its most typical manifestation was the ancient-rooted notion of fortuna. However, the fortune concept lost its explanatory power during the 17th century, in connection with scientific, economic and social changes. There are many signs that, at the same time, ideas about divine providence were transformed. Although there are signs of this already in the 16th century (such as in the popular Fortunatus), especially from the end of the 17th century onwards there is a noticeable erosion of the traditional metaphysics based on the centrality of the providential God. Literary studies can also contribute to the study of changing ideas about providence, since it is often in genres that provide a more flexible framework than theological or philosophical discourse that the first signs of change appear. Miklós Bethlen’s autobiography, written in the early 18th century is a good example of this. The overall framework of the self-interpretation of the autobiography is a belief in providence, yet there are also elements that are in tension with it. These have so far only been referred to in a few scattered references in the literature, but have not been comprehensively examined and interpreted. An examination of the literary conceptualisation of providence and chance can show how traditional conceptual frameworks become problematic in the face of new experiences, while their meaning is also modified.

Ghada AlNemer, Mohammad Farid, Rehan Ali

This work presents an extragradient-type iterative process combined with the viscosity method to find a common solution to a split generalized variational-like inequality, a variational inequality, and a fixed point problem associated with a family of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>ε</mi></semantics></math></inline-formula>-strict pseudo-contractive mappings and a nonexpansive operator in Hilbert spaces. Strong convergence of the proposed algorithm is established, with some remarks derived from the main theorem. Numerical experiments are carried out to verify the applicability of the method and provide comparative observations. The results broaden and unify a range of existing contributions in this field.

Khaled Abud, Sergey Lavrushkin, Dmitry Vatolin

Neural-network-based models have made considerable progress in many computer vision areas over recent years. However, many works have exposed their vulnerability to malicious input data manipulation—that is, to adversarial attacks. Although many recent works have thoroughly examined the adversarial robustness of classifiers, the robustness of Image Quality Assessment (IQA) methods remains understudied. This paper addresses this gap by proposing FM-GOAT (Frequency-Masked Gradient Orthogonalization Attack), a novel white box adversarial method tailored for no-reference IQA models. Using a novel gradient orthogonalization technique, FM-GOAT uniquely optimizes adversarial perturbations against multiple perceptual constraints to minimize visibility, moving beyond traditional <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>l</mi><mi>p</mi></msub></semantics></math></inline-formula>-norm bounds. We evaluate FM-GOAT on seven state-of-the-art NR-IQA models across three image and video datasets, revealing significant vulnerability to the proposed attack. Furthermore, we examine the applicability of adversarial purification methods to the IQA task, as well as their efficiency in mitigating white box adversarial attacks. By studying the activations from models’ intermediate layers, we explore their behavioral patterns in adversarial scenarios and discover valuable insights that may lead to better adversarial detection.

Shaull Almagor, Neta Dafni, Ishai Salgado

Jumping automata are finite automata that read their input in a non-sequential manner, by allowing a reading head to ``jump'' between positions on the input, consuming a permutation of the input word. We argue that allowing the head to jump should incur some cost. To this end, we propose four quantitative semantics for jumping automata, whereby the jumps of the head in an accepting run define the cost of the run. The four semantics correspond to different interpretations of jumps: the \emph{absolute distance} semantics counts the distance the head jumps, the \emph{reversal} semantics counts the number of times the head changes direction, the \emph{Hamming distance} measures the number of letter-swaps the run makes, and the \emph{maximum jump} semantics counts the maximal distance the head jumps in a single step, We study these measures, with the main focus being the \emph{boundedness problem}: given a jumping automaton, decide whether its (quantitative) language is bounded by some given number $k$. We establish the decidability and complexity for this problem under several variants.

Wei Ye, Xinan He, Feng Ding

In this paper, we propose a novel method for detecting DeepFakes, enhancing the generalization of detection through semantic decoupling. There are now multiple DeepFake forgery technologies that not only possess unique forgery semantics but may also share common forgery semantics. The unique forgery semantics and irrelevant content semantics may promote over-fitting and hamper generalization for DeepFake detectors. For our proposed method, after decoupling, the common forgery semantics could be extracted from DeepFakes, and subsequently be employed for developing the generalizability of DeepFake detectors. Also, to pursue additional generalizability, we designed an adaptive high-pass module and a two-stage training strategy to improve the independence of decoupled semantics. Evaluation on FF++, Celeb-DF, DFD, and DFDC datasets showcases our method's excellent detection and generalization performance. Code is available at: https://github.com/leaffeall/DFS-GDD.

Larisa Latypova, Fadis Murzakhanov, George Mamin et al.

The distinct spin, optical, and coherence characteristics of solid-state spin defects in semiconductors have positioned them as potential qubits for quantum technologies. Both bulk and two-dimensional materials, with varying structural properties, can serve as crystalline hosts for color centers. In this study, we conduct a comparative analysis of the spin–optical, electron–nuclear, and relaxation properties of nitrogen-bound vacancy defects using electron paramagnetic resonance (EPR) and electron–nuclear double resonance (ENDOR) techniques. We examine key parameters of the spin Hamiltonian for the nitrogen vacancy (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mrow><mi>N</mi><mi>V</mi></mrow><mrow><mo>−</mo></mrow></msup></mrow></semantics></math></inline-formula>) center in 4H-SiC: <i>D</i> = 1.3 GHz, <i>A_zz</i> = 1.1 MHz, and <i>C</i>_Q = 2.53 MHz, as well as for the boron vacancy (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msubsup><mrow><mi>V</mi></mrow><mrow><mi>B</mi></mrow><mrow><mo>−</mo></mrow></msubsup></mrow></semantics></math></inline-formula>) in hBN: <i>D</i> = 3.6 GHz, <i>A_zz</i> = 85 MHz, and <i>C</i>_Q = 2.11 MHz, and their dependence on the material matrix. The spin–spin relaxation times <i>T</i>₂ (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mrow><mi>N</mi><mi>V</mi></mrow><mrow><mo>−</mo></mrow></msup></mrow></semantics></math></inline-formula> center: 50 µs and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msubsup><mrow><mi>V</mi></mrow><mrow><mi>B</mi></mrow><mrow><mo>−</mo></mrow></msubsup></mrow></semantics></math></inline-formula>: 15 µs) are influenced by the local nuclear environment and spin diffusion while Rabi oscillation damping times depend on crystal size and the spatial distribution of microwave excitation. The ENDOR absorption width varies significantly among color centers due to differences in crystal structures. These findings underscore the importance of selecting an appropriate material platform for developing quantum registers based on high-spin color centers in quantum information systems.

Zhongtian Dong, Marçal Comajoan Cara, Gopal Ramesh Dahale et al.

This paper presents a comparative analysis of the performance of Equivariant Quantum Neural Networks (EQNNs) and Quantum Neural Networks (QNNs), juxtaposed against their classical counterparts: Equivariant Neural Networks (ENNs) and Deep Neural Networks (DNNs). We evaluate the performance of each network with three two-dimensional toy examples for a binary classification task, focusing on model complexity (measured by the number of parameters) and the size of the training dataset. Our results show that the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi mathvariant="double-struck">Z</mi><mn>2</mn></msub><mo>×</mo><msub><mi mathvariant="double-struck">Z</mi><mn>2</mn></msub></mrow></semantics></math></inline-formula> EQNN and the QNN provide superior performance for smaller parameter sets and modest training data samples.

Furkan Dikmen, Elena Guerzoni, Ömer Demirok

Yun Xing, Jian Kang, Aoran Xiao et al.

Vision-Language Pre-training has demonstrated its remarkable zero-shot recognition ability and potential to learn generalizable visual representations from language supervision. Taking a step ahead, language-supervised semantic segmentation enables spatial localization of textual inputs by learning pixel grouping solely from image-text pairs. Nevertheless, the state-of-the-art suffers from clear semantic gaps between visual and textual modality: plenty of visual concepts appeared in images are missing in their paired captions. Such semantic misalignment circulates in pre-training, leading to inferior zero-shot performance in dense predictions due to insufficient visual concepts captured in textual representations. To close such semantic gap, we propose Concept Curation (CoCu), a pipeline that leverages CLIP to compensate for the missing semantics. For each image-text pair, we establish a concept archive that maintains potential visually-matched concepts with our proposed vision-driven expansion and text-to-vision-guided ranking. Relevant concepts can thus be identified via cluster-guided sampling and fed into pre-training, thereby bridging the gap between visual and textual semantics. Extensive experiments over a broad suite of 8 segmentation benchmarks show that CoCu achieves superb zero-shot transfer performance and greatly boosts language-supervised segmentation baseline by a large margin, suggesting the value of bridging semantic gap in pre-training data.

Juha Kontinen, Max Sandström, Jonni Virtema

We introduce and develop a set-based semantics for asynchronous TeamLTL. We consider two canonical logics in this setting: the extensions of TeamLTL by the Boolean disjunction and by the Boolean negation. We establish fascinating connections between the original semantics based on multisets and the new set-based semantics as well as show one of the first positive complexity theoretic results in the temporal team semantics setting. In particular we show that both logics enjoy normal forms that can be utilised to obtain results related to expressivity and complexity (decidability) of the new logics. We also relate and apply our results to recently defined logics whose asynchronicity is formalized via time evaluation functions.

Leonardo Pacheco

We define game semantics for the constructive $μ$-calculus and prove its equivalence to bi-relational semantics. As an application, we use the game semantics to prove that the $μ$-calculus collapses to modal logic over the modal logic $\mathsf{IS5}$. We then show the completeness of $\mathsf{IS5}$ extended with fixed-point operators.

Nelson J. G. Fonseca, Sophie-Abigaël Gomanne, José Rico-Fernández et al.

This paper reports on the design and experimental validation of a fully-metallic double-ridged waveguide 10 × 10 Rotman lens additively manufactured as a single part. The wide band operation of this quasi-optical beamformer enables us to cover the uplink and downlink frequencies allocated to satellite communications in the K/K<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mrow></mrow><mi mathvariant="normal">a</mi></msub></semantics></math></inline-formula>-band, from 17.3 GHz to 30 GHz. The feeding port design was adjusted to enable vertical printing, thus minimizing the use of supporting structures. A prototype was manufactured and tested. The reported results indicate losses in the range of 0.5 dB in the lower-frequency band and 0.8 dB in the upper-frequency band, including the waveguide transitions added for test purposes. The measured reflection and coupling coefficients remain below −11.5 dB over the operating band. The standard deviation of the residual phase error across the array ports is below <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>5</mn><mo>°</mo></mrow></semantics></math></inline-formula> in simulation and below <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>10</mn><mo>°</mo></mrow></semantics></math></inline-formula> in measurements. Array factors synthesized using the scattering parameters confirm the good stability of the beamforming functionality over the wide frequency band analyzed. This monolithic design is a promising step toward more integrated antenna systems, such as a compact dual-stack configuration for planar array design.

Jordanoska Izabela, Kocher Anna, Bendezú-Araujo Raúl

This special issue focuses on the theoretical and empirical underpinnings of truth-marking. The names that have been used to refer to this phenomenon include, among others, counter-assertive focus, polar(ity) focus, verum focus, emphatic polarity or simply verum. This terminological variety is suggestive of the wide range of ideas and conceptions that characterizes this research field. This collection aims to get closer to the core of what truly constitutes verum. We want to expand the empirical base and determine the common and diverging properties of truth-marking in the languages of the world. The objective is to set a theoretical and empirical baseline for future research on verum and related phenomena.

Dim Idrisov, Petr Parfenov, Arkadiy Taranenko

The elliptic flow (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>v</mi><mn>2</mn></msub></semantics></math></inline-formula>) of produced particles is one of the important observables sensitive to the transport properties of the strongly interacting matter created in relativistic heavy-ion collisions. Detailed differential measurements of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>v</mi><mn>2</mn></msub></semantics></math></inline-formula> are also foreseen in the future Multi-Purpose Detector (MPD) experiment at the Nuclotron based Ion Collider fAcility (NICA) at collision energies <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msqrt><msub><mi>s</mi><mrow><mi>N</mi><mi>N</mi></mrow></msub></msqrt></semantics></math></inline-formula> = 4–11 GeV. Elliptic flow strongly depends on collision geometry, defined by the impact parameter <i>b</i>. Usually <i>b</i> is an input to theoretical calculations and can be deduced from experimental observables in the final state using the centrality procedure. In this work, we investigate the influence of the choice of centrality procedure on the elliptic flow measurements at NICA energies.