Hasil untuk "Semantics"

A. Ramesh, Prafulla Dhariwal, Alex Nichol et al.

Contrastive models like CLIP have been shown to learn robust representations of images that capture both semantics and style. To leverage these representations for image generation, we propose a two-stage model: a prior that generates a CLIP image embedding given a text caption, and a decoder that generates an image conditioned on the image embedding. We show that explicitly generating image representations improves image diversity with minimal loss in photorealism and caption similarity. Our decoders conditioned on image representations can also produce variations of an image that preserve both its semantics and style, while varying the non-essential details absent from the image representation. Moreover, the joint embedding space of CLIP enables language-guided image manipulations in a zero-shot fashion. We use diffusion models for the decoder and experiment with both autoregressive and diffusion models for the prior, finding that the latter are computationally more efficient and produce higher-quality samples.

Matthew E. Peters, Mark Neumann, Mohit Iyyer et al.

We introduce a new type of deep contextualized word representation that models both (1) complex characteristics of word use (e.g., syntax and semantics), and (2) how these uses vary across linguistic contexts (i.e., to model polysemy). Our word vectors are learned functions of the internal states of a deep bidirectional language model (biLM), which is pre-trained on a large text corpus. We show that these representations can be easily added to existing models and significantly improve the state of the art across six challenging NLP problems, including question answering, textual entailment and sentiment analysis. We also present an analysis showing that exposing the deep internals of the pre-trained network is crucial, allowing downstream models to mix different types of semi-supervision signals.

V. Dijk

E. Clark

J. Coates

G. Kahn

Gennaro Chierchia, Sally Mcconnell-Ginet

A. Wierzbicka

Abdulaziz Khalid Alsharidi, Ahmet Bekir

The paper consists of various types of wave solutions for the truncated M-fractional Bateman–Burgers equation, a significant mathematical physics equation. This model describes the nonlinear waves and solitons in different physical fields such as optical fibers, plasma physics, fluid dynamics, traffic flow, etc. Through the application of the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mo form="prefix">exp</mo><mi>a</mi></msub></semantics></math></inline-formula> function method and the modified simplest equation method, we are able to obtain exact series of soliton solutions. The results differ from the current solutions of the Bateman–Burgers model because of the fractional derivative. The achieved results could be helpful in various engineering and scientific domains. The Mathematica software is used to assist in obtaining and verifying the exact solutions and to obtain contour plots of the solutions in two and three dimensions. To ensure that the model in question is stable, a stability analysis is also carried out using the modulation instability method. Future research on the system in question and related systems will benefit from the findings. The methods used are simple and effective.

Yongbin Tan, Hong Wang, Rongfeng Cai et al.

Unsupervised representation learning can train deep learning models to formally express the semantic connotations of objects in the case of unlabeled data, which can effectively realize the expression of the semantics of geospatial entity categories in application scenarios lacking expert knowledge and help achieve the deep fusion of geospatial data. In this paper, a method for the semantic representation of the geospatial entity categories (denoted as feature embedding) is presented, taking advantage of the characteristic that regions with similar distributions of geospatial entity categories also have a certain level of similarity. To construct the entity category embedding, a spatial proximity graph of entities and an adjacency matrix of entity categories are created using a large number of geospatial entities obtained from OSM (OpenStreetMap). The cosine similarity algorithm is then employed to measure the similarity between these embeddings. Comparison experiments are then conducted by comparing the similarity results from the standard model. The results show that the results of this model are basically consistent with the standard model (Pearson correlation coefficient = 0.7487), which verifies the effectiveness of the feature embedding extracted by this model. Based on this, this paper applies the feature embedding to the regional similarity task, which verifies the feasibility of using the model in the downstream task. It provides a new idea for realizing the formal expression of the unsupervised entity category semantics.

Wenhao Cui, Zhaoxin Wang, Lei Ma

Traditional emotion decoding methods typically rely on short sequences with limited context and coarse-grained emotion categories. To address these limitations, we proposed the Semantic and Emotion Decoding Generative Pre-trained Transformer (SED-GPT), a non-invasive method for long-sequence fine-grained semantics and emotions decoding on extended narrative stimuli. Using a publicly available fMRI dataset from 8 participants, this exploratory study investigates the feasibility of reconstructing complex semantic and emotional states from brain activity. SED-GPT achieves a BERTScore-F1 of 0.650 on semantic decoding and attains a cosine similarity (CS) of 0.504 and a Jensen–Shannon similarity (JSS) of 0.469 for emotion decoding (<i>p</i> < 0.05). Functional connectivity analyses reveal persistent coupling between the language network and the emotion network, which provides neural evidence for the language–emotion interaction mechanism in Chinese. These findings should be interpreted as pilot-level feasibility evidence.

Jutarut Chaoraingern, Arjin Numsomran

The accurate real-time estimation of the remaining useful life (RUL) of lithium-polymer (LiPo) batteries is a critical enabler for ensuring the safety, reliability, and operational efficiency of unmanned aerial vehicles (UAVs). Nevertheless, achieving such prognostics on resource-constrained embedded platforms remains a considerable technical challenge. This study proposes an end-to-end TinyML-based framework that integrates embedded sensor data fusion with an optimized feedforward neural network (FFNN) model for efficient RUL estimation under strict hardware limitations. The system collects voltage, discharge time, and capacity measurements through a lightweight data fusion pipeline and leverages the Edge Impulse platform with the EON™Compiler for model optimization. The trained model is deployed on a dual-core ARM Cortex-M0+ Raspberry Pi RP2040 microcontroller, communicating wirelessly with a LabVIEW-based visualization system for real-time monitoring. Experimental validation on an 80-gram UAV equipped with a 1100 mAh LiPo battery demonstrates a mean absolute error (<i>MAE</i>) of 3.46 cycles and a root mean squared error (<i>RMSE</i>) of 3.75 cycles. Model testing results show an overall accuracy of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>98.82</mn><mo>%</mo></mrow></semantics></math></inline-formula>, with a mean squared error (<i>MSE</i>) of 55.68, a mean absolute error (<i>MAE</i>) of 5.38, and a variance score of 0.99, indicating strong regression precision and robustness. Furthermore, the quantized (int8) version of the model achieves an inference latency of 2 ms, with memory utilization of only 1.2 KB RAM and 11 KB flash, confirming its suitability for real-time deployment on resource-constrained embedded devices. Overall, the proposed framework effectively demonstrates the feasibility of combining embedded sensor data fusion and TinyML to enable accurate, low-latency, and resource-efficient real-time RUL estimation for UAV battery health management.

Jerzy Smardzewski, Michał Słonina

The cores of honeycomb panels are usually made of hexagonal cells. Due to their structure, they create anticlastic surfaces that are difficult to use in furniture design. Synclastic surfaces in lightweight sandwich panels are typically associated with auxetic cores characterized by a negative Poisson’s ratio. This study aimed to transform the hexagonal cell cores into cells with a negative or positive Poisson’s ratio (NPR, PPR), enabling these cores to form synclastic surfaces. New core structures for synclastic furniture sandwich honeycomb panels were modeled numerically and experimentally. It has been demonstrated that reentrant cells with NPR create synclastic surfaces, and new shapes of core cells, created by transforming hexagonal cells with PPR, also enable the formation of synclastic surfaces. Cores’ synclasticity was assessed in two orthogonal planes using physical models and Finite Element Analysis (FEA). A new and original discovery is the demonstration that not only auxetic but also modified hexagonal cells with Poisson’s ratios of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mrow><mi>ν</mi></mrow><mrow><mi>x</mi><mi>y</mi></mrow></msub></mrow></semantics></math></inline-formula> = 0.545 and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mrow><mi>ν</mi></mrow><mrow><mi>y</mi><mi>x</mi></mrow></msub></mrow></semantics></math></inline-formula> = 0.512, respectively, exhibit excellent synclastic properties. The agreement between FEA and experiment was very high. The results show that not only NPR but also cell topology provides a practical route to the synclastic formation of cores without the use of auxetic materials.

Manuel G. Pinedo-Cuba, José M. Caicedo-Roque, Jessica Padilla-Pantoja et al.

Magnetic shape memory alloys have attracted considerable attention due to their multifunctional properties. Among these materials, Ni-Mn-Ga alloys are distinguished by their ability to achieve up to 10% strain when exposed to a magnetic field, a characteristic predominantly observed in single-crystal samples. Consequently, it is essential to develop nanomaterials with a crystal structure closely resembling that of a single crystal. In this study, an epitaxial Ni-Mn-Ga thin film was fabricated using Pulsed Laser Deposition on an Al₂O₃ <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>(</mo><mn>11</mn><mover accent="true"><mn>2</mn><mo stretchy="false">¯</mo></mover><mn>0</mn><mo>)</mo></mrow></semantics></math></inline-formula> single-crystal substrate. The crystal structure was characterised through X-ray diffraction methodologies, such as symmetrical <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>2</mn><mi>θ</mi><mo>−</mo><mi>ω</mi></mrow></semantics></math></inline-formula> scans, pole figures, and reciprocal space maps. The results indicated that the sample was mainly in a slightly distorted cubic austenite phase, and some incipient martensite phase also appeared. A detailed microstructural analysis, performed by transmission electron microscopy, confirmed that certain regions of the sample exhibited an incipient transformation to the martensite phase. Regions closer to the substrate retained the austenite phase, suggesting that the constraint imposed by the substrate inhibits the phase transition. These results indicate that it is possible to grow high crystalline quality thin films of Ni-Mn-Ga by Pulsed Laser Deposition.

Junping Li, Mixuan Hou

Let <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>{</mo><msub><mi>Z</mi><mi>n</mi></msub><mo>:</mo><mi>n</mi><mo>≥</mo><mn>0</mn><mo>}</mo></mrow></semantics></math></inline-formula> be a Galton–Watson system with a circular mechanism <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi mathvariant="bold-italic">a</mi><mo>∗</mo><mi mathvariant="bold-italic">b</mi></mrow></semantics></math></inline-formula>, where <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi mathvariant="bold-italic">a</mi><mo>=</mo><msubsup><mrow><mo>{</mo><msub><mi>a</mi><mi>j</mi></msub><mo>}</mo></mrow><mrow><mi>j</mi><mo>=</mo><mn>0</mn></mrow><mo>∞</mo></msubsup></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi mathvariant="bold-italic">b</mi><mo>=</mo><msubsup><mrow><mo>{</mo><msub><mi>b</mi><mi>j</mi></msub><mo>}</mo></mrow><mrow><mi>j</mi><mo>=</mo><mn>0</mn></mrow><mo>∞</mo></msubsup></mrow></semantics></math></inline-formula> are probability distributions on <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi mathvariant="bold">Z</mi><mo>+</mo></msub><mo>:</mo><mo>=</mo><mrow><mo>{</mo><mn>0</mn><mo>,</mo><mn>1</mn><mo>,</mo><mn>2</mn><mo>,</mo><mo>⋯</mo><mo>}</mo></mrow></mrow></semantics></math></inline-formula>. Let <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>m</mi><mi>a</mi></msub><mo>:</mo><mo>=</mo><mstyle displaystyle="true"><munderover><mo>∑</mo><mrow><mi>j</mi><mo>=</mo><mn>0</mn></mrow><mo>∞</mo></munderover></mstyle><mi>j</mi><msub><mi>a</mi><mi>j</mi></msub></mrow></semantics></math></inline-formula>, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>m</mi><mi>b</mi></msub><mo>:</mo><mo>=</mo><mstyle displaystyle="true"><munderover><mo>∑</mo><mrow><mi>j</mi><mo>=</mo><mn>0</mn></mrow><mo>∞</mo></munderover></mstyle><mi>j</mi><msub><mi>b</mi><mi>j</mi></msub></mrow></semantics></math></inline-formula>. The extinction property of such branching systems is first studied. Then, it is proved that there exists <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>Γ</mi><mi>n</mi></msub></semantics></math></inline-formula> such that <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>W</mi><mi>n</mi></msub><mo>=</mo><msubsup><mi>Γ</mi><mi>n</mi><mrow><mo>−</mo><mn>1</mn></mrow></msubsup><msub><mi>Z</mi><mi>n</mi></msub></mrow></semantics></math></inline-formula> is an integrable martingale and hence converges to some random variable <i>W</i>. Moreover, for the case that <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>a</mi><mn>0</mn></msub><mo>=</mo><msub><mi>b</mi><mn>0</mn></msub><mo>=</mo><mn>0</mn><mo>,</mo><mo> </mo><msub><mi>a</mi><mn>1</mn></msub><mo>,</mo><msub><mi>b</mi><mn>1</mn></msub><mo>∈</mo><mrow><mo>(</mo><mn>0</mn><mo>,</mo><mn>1</mn><mo>)</mo></mrow></mrow></semantics></math></inline-formula>, the convergence rates to 0 of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>P</mi><mfenced separators="" open="(" close=")"><mfenced separators="" open="|" close="|"><mstyle scriptlevel="0" displaystyle="true"><mfrac><msub><mi>Z</mi><mrow><mi>n</mi><mo>+</mo><mn>1</mn></mrow></msub><msub><mi>Z</mi><mi>n</mi></msub></mfrac></mstyle><mo>−</mo><msub><mi>ω</mi><mi>n</mi></msub></mfenced><mo>></mo><mi>ε</mi><mo>∣</mo><msub><mi>Z</mi><mn>0</mn></msub><mo>=</mo><mn>1</mn></mfenced><mrow><mo>,</mo><mi>P</mi><mo>(</mo><mo>|</mo></mrow><msub><mi>W</mi><mi>n</mi></msub><mo>−</mo><mi>W</mi><mrow><mo>|</mo><mo>></mo><mi>ε</mi><mo>∣</mo><msub><mi>Z</mi><mn>0</mn></msub><mo>=</mo><mn>1</mn><mo>)</mo></mrow></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>P</mi><mfenced separators="" open="(" close=")"><mfenced separators="" open="|" close="|"><mstyle scriptlevel="0" displaystyle="true"><mfrac><msub><mi>Z</mi><mrow><mi>n</mi><mo>+</mo><mn>1</mn></mrow></msub><msub><mi>Z</mi><mi>n</mi></msub></mfrac></mstyle><mo>−</mo><msub><mi>ω</mi><mi>n</mi></msub></mfenced><mo>></mo><mi>ε</mi><mo>∣</mo><mi>W</mi><mo>≥</mo><mi>δ</mi><mo>,</mo><msub><mi>Z</mi><mn>0</mn></msub><mo>=</mo><mn>1</mn></mfenced></mrow></semantics></math></inline-formula> as <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>n</mi><mo>→</mo><mo>∞</mo></mrow></semantics></math></inline-formula> are presented for <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>ε</mi><mo>,</mo><mi>δ</mi><mo>></mo><mn>0</mn></mrow></semantics></math></inline-formula> under various moment conditions on <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>{</mo><msub><mi>a</mi><mi>j</mi></msub><mo>}</mo></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>{</mo><msub><mi>b</mi><mi>j</mi></msub><mo>}</mo></mrow></semantics></math></inline-formula>, where <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>ω</mi><mi>n</mi></msub><mo>=</mo><msub><mi>m</mi><mi>a</mi></msub></mrow></semantics></math></inline-formula> or <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>m</mi><mi>b</mi></msub></semantics></math></inline-formula> for <i>n</i> being even or odd, respectively. It is further shown that the first rate is geometric while the last two rates are supergeometric under a finite moment generating function hypothesis.

Andreas Lööw

Despite numerous previous formalisation projects targeting Verilog, the semantics of Verilog defined by the Verilog standard -- Verilog's simulation semantics -- has thus far eluded definitive mathematical formalisation. Previous projects on formalising the semantics have made good progress but no previous project provides a formalisation that can be used to execute or formally reason about real-world hardware designs. In this paper, we show that the reason for this is that the Verilog standard is inconsistent both with Verilog practice and itself. We pinpoint a series of problems in the Verilog standard that we have identified in how the standard defines the semantics of the subset of Verilog used to describe hardware designs, that is, the synthesisable subset of Verilog. We show how the most complete Verilog formalisation to date inherits these problems and how, after we repair these problems in an executable implementation of the formalisation, the repaired implementation can be used to execute real-world hardware designs. The existing formalisation together with the repairs hence constitute the first formalisation of Verilog's simulation semantics compatible with real-world hardware designs. Additionally, to make the results of this paper accessible to a wider (nonmathematical) audience, we provide a visual formalisation of Verilog's simulation semantics.

Kobus Barnard, D. Forsyth

Andrew Chio, Jian Peng, Nalini Venkatasubramanian

Urban communities rely on built utility infrastructures as critical lifelines that provide essential services such as water, gas, and power, to sustain modern socioeconomic systems. These infrastructures consist of underground and surface-level assets that are operated and geo-distributed over large regions where continuous monitoring for anomalies is required but challenging to implement. This article addresses the problem of deploying heterogeneous Internet of Things sensors in these networks to support future decision-support tasks, for example, anomaly detection, source identification, and mitigation. We use stormwater as a driving use case; these systems are responsible for drainage and flood control, but act as conduits that can carry contaminants to the receiving waters. Challenges toward effective monitoring include the transient and random nature of the pollution incidents, the scarcity of historical data, the complexity of the system, and technological limitations for real-time monitoring. We design a SemanTics-aware sEnsor Placement framework (STEP) to capture pollution incidents using structural, behavioral, and semantic aspects of the infrastructure. We leverage historical data to inform our system with new, credible instances of potential anomalies. Several key topological and empirical network properties are used in proposing candidate deployments that optimize the balance between multiple objectives. We also explore the quality of anomaly representation in the network through new perspectives, and provide techniques to enhance the realism of the anomalies considered in a network. We evaluate STEP on six real-world stormwater networks in Southern California, USA, which shows its efficacy in monitoring areas of interest over other baseline methods.

Vasileios Koutavas, Yu-Yang Lin, Nikos Tzevelekos

We present a big-step and small-step operational semantics for Yul -- the intermediate language used by the Solidity compiler to produce EVM bytecode -- in a mathematical notation that is congruous with the literature of programming languages, lends itself to language proofs, and can serve as a precise, widely accessible specification for the language. Our two semantics stay faithful to the original, informal specification of the language but also clarify under-specified cases such as void function calls. Our presentation allows us to prove the equivalence between the two semantics. We also implement the small-step semantics in an interpreter for Yul which avails of optimisations that are provably correct. We have tested the interpreter using tests from the Solidity compiler and our own. We envisage that this work will enable the development of verification and symbolic execution technology directly in Yul, contributing to the Ethereum security ecosystem, as well as aid the development of a provably sound future type system.

Daniel Quigley

This paper proves a homomorphism between extensional formal semantics and distributional vector space semantics, demonstrating structural compatibility. Formal semantics models meaning as reference, using logical structures to map linguistic expressions to truth conditions, while distributional semantics represents meaning through word vectors derived from contextual usage. By constructing injective mappings that preserve semantic relationships, we show that every semantic function in an extensional model corresponds to a compatible vector space operation. This result respects compositionality and extends to function compositions, constant interpretations, and $n$-ary relations. Rather than pursuing unification, we highlight a mathematical foundation for hybrid cognitive models that integrate symbolic and sub-symbolic reasoning and semantics. These findings support multimodal language processing, aligning `meaning as reference' (Frege, Tarski) with `meaning as use' (Wittgenstein, Firth).