Search Results (1,327)

Accurate wireless positioning has remained challenging under dynamic bandwidth conditions and outdoor multipath environments that are typical in Internet of Things (IoT) and autonomous aerial vehicle (AAV) applications. Conventional learning-based localization methods rely on bandwidth-specific channel state information (CSI) representations, which causes the trained models to be inapplicable or less adaptive when the signal bandwidth differs from that used during training. To overcome this limitation, a unified and neural network-oriented framework is proposed, which constructs bandwidth-adaptive power delay profile (PDP) representations for learning-based models. A PDP preprocessing scheme through adaptive zero-padding and oversampled IFFT of heterogeneous CSI is introduced to generate dimension-consistent and delay-aligned neural network inputs. To enhance robustness, a sub-band-sliced PDP representation is developed to enhance model robustness, where each bandwidth is divided into equal-width sub-bands whose PDPs are independently processed and organized as Transformer tokens. A dedicated Transformer is designed to get the location estimation from PDPs of multi-access points. Simulation results have demonstrated that the proposed preprocessing-PDP-plus-Transformer framework achieves superior cross-bandwidth generalization and localization accuracy, compared to analytical and learning-based baselines. Full article

This paper is a focused, standards-aligned survey of where autoregressive (AR) large language models (LLMs) tend to break down when deployed inside industrial informatics workflows that must satisfy long-horizon objectives, hard constraints, traceability, and functional-safety obligations (e.g., IEC 61508/ISO 26262/ISO 21448). Rather than claiming new algorithms or experiments, we synthesize and organize prior work into (i) a control-oriented taxonomy of four AR failure modes that recur in practice (compounding error, myopic objectives, data brittleness/hallucinations, and scaling/latency inefficiencies), (ii) a catalog of standards-compatible deployment patterns that mitigate these issues (human-gated LLM-in-the-loop, retrieval + verification pipelines, planner-of-record architectures, and runtime assurance envelopes), and (iii) an operational decision framework (criteria table with observable proxies, a stepwise decision procedure, and worked examples) for deciding when token-centric mitigations are sufficient versus when state/world-model components become warranted. Joint Embedding Predictive Architectures (JEPA) and Hierarchical JEPA (H-JEPA) JEPA are proposed as representative state-predictive architectures, with discussion explicitly bounded by currently available empirical evidence; we explicitly note that the published evidence base is currently concentrated on vision/multimodal benchmarks and that industrial control validation remains limited. To make evidence boundaries transparent, we introduce (a) a survey method (scope, inclusion/exclusion criteria, and data-extraction fields), (b) a comparison matrix across representative prior systems, and (c) an evidence map that links each deployment pattern to peer-reviewed empirical findings and system reports. Full article

One major challenge faced by deep learning-based methods that detect target objects in the form of bounding boxes is object occlusion. High degrees of occlusion significantly diminish the accuracy of instance segmentation. Nonetheless, complex-valued Fourier descriptors can robustly represent object boundaries using minimal information. In this study, the impact of integrating Fourier descriptors—renowned for their strong representational capacity—with deep network models (UNet) that exhibit high generalization performance on instance segmentation accuracy was investigated. Within the scope of the research, nine network models were designed based on different strategies for utilizing frequency components. These variants fall into four strategy families: (i) UNet-style spectrum regression on fixed low-frequency windows (FUNet), (ii) magnitude-guided frequency selection/ROI construction (FUNet–Thr, FUNet–BBox), (iii) sequence models over tokenized FFT coefficients (BiLSTM Patch/Sorted), and (iv) encoder-only spectrum predictors with different depth/capacity (EncoderFFT1/2). To fairly evaluate the models’ performance in segmenting objects subjected to disruptive factors (e.g., occlusion, blurring, noise), a specialized synthetic dataset was prepared. The task is formulated as single-target (single-instance), single-class segmentation. This dataset, automatically generated according to initial parameter values, contains images of objects moving at various speeds within a single frame. Among these models, the one termed FUNet, which relies on partial matching of central frequency components, achieved the highest segmentation accuracy despite the disruptive effects. Under the challenging Dataset 8 setting, the proposed FUNet achieved the highest overlap-based performance (Dice = 0.9329, IoU = 0.8842) among Attention U-Net, U-Net, and FourierNet, with statistically significant gains confirmed by paired per-image tests. Full article

Background/Objectives: Predicting diseases based on the gut microbiome pattern is still difficult because of compositional shortcomings, batch heterogeneity, and scanty modeling of inter-taxon interactions. This study introduces a Dysbiosis-Aware Multiset Transformer Framework called DysbioFormer, which predicts state diseases by recognizing patterns of microbes. Methods: The current methods are mainly based on flat abundance representations or fixed-order models which limit the capability of describing intricate interactions of communities and evolutionary structure. Results: DysbioFormer is a solution to these shortcomings, in which each sample of the microbiome is modeled as a permutation-invariant multiset of taxonomic tokens with compositional, phylogenetic, and harmonized cohort data. Stacked Set Attention Blocks are used to learn relational dependencies between taxa, whereas Pooling-by-Multihead-Attention is used to aggregate global disease-level embeddings and this is not based on sequence assumptions. The model has been tested on MicrobiomeHD, which consists of a wide variety of human gut microbiome samples at a variety of disease conditions and healthy controls. Experimental results demonstrate strong diagnostic performance, achieving an accuracy of 97%, an AUC of 0.97, and an F1-score of 96%, consistently outperforming classical machine learning models under identical evaluation protocols. Attention-derived signatures also can give interpretable connections among predictive results and disease-linked microbial taxa, enhancing biological plausibility. Conclusions: The suggested architecture enables scalable, cohort-agnostic microbial diagnostics, and provides a principled route to transforming the complex information of the microbiome into reliable clinical information. DysbioFormer creates a universal basis of future microbiome-based disease screening and precision health uses. Its design allows extending towards multi-omics integration, longitudinal studies, and decision-support infrastructure, supporting microbiome-informed translational medicine in a variety of clinical research settings. Full article

Super-resolution technology significantly enhances the visual quality of low-resolution medical images, resulting in ultra-high-resolution clear images. Super-resolution technology based on artificial intelligence has achieved great success in reconstruction quality. However, like the image restoration task, super-resolution is also an ill-posed problem, and current work lacks consideration of trustworthiness. Medical image super-resolution needs to ensure clarity and, more importantly, to ensure that the output image is reliable and does not produce false details and mislead the diagnosis. To address the trustworthy issue of medical image super-resolution, we design a novel hybrid loss that combines a hinge-based adversarial term with a PSNR-based regularization. In the designed loss function, the adversarial term makes the reconstructed result close to the distribution of the true high-resolution image, thus generating more refined high-frequency textures, while the PSNR-based regularization term explicitly reduces the deviation from the ground truth. We apply this loss in the global-token U-Net backbone network and add a lightweight VGG as the discriminator for adversarial terms. We empirically verify that integrating the proposed methods can enhance the trustworthiness of medical image super-resolution technology while maintaining high reconstruction quality. Full article

The prevalence of ground fissures in deformation-affected areas has intensified, presenting serious risks to both operational safety and the local natural environment. Fissures in these disturbed terrains are typically characterized by elongated morphologies and large-scale variations, which pose substantial challenges to accurate feature extraction. To address these complexities, this paper proposes a semantic segmentation network termed MGF-UNet. In the shallow layers, we integrate multi-scale feature sensing (MFS) and grouped efficient multi-scale attention (EMA) to sharpen anisotropic textures and boundary details under high-resolution representations. For the deeper layers, a Token-Selective Context Transformer (TSCT) is designed to perform selective global modeling on high-level semantic features, effectively capturing long-range dependencies while preserving the structural integrity of elongated fissures. Meanwhile, we employ feature-wise linear modulation (FiLM) to derive pixel-wise affine parameters from shallow structures, which pre-modulate deep features and strengthen cross-level interactions. In the decoder, a Fourier transform-based adaptive feature fusion (AFF) module suppresses background noise and enhances boundary contrast, followed by cross-scale aggregation for final prediction.Benchmark tests conducted on the mining-area fissure dataset (MFD) and road-based datasets demonstrate that MGF-UNet achieves an accuracy of 78.2%, a Dice score of 81.4%, and an IoU of 68.6%, outperforming existing mainstream networks. The results confirm that MGF-UNet provides an effective solution for automatic fissure extraction in deformation-prone environments, offering significant potential for geohazard monitoring and ecological restoration. Full article

This paper presents BPMN Assistant, a tool that leverages Large Language Models for natural language-based creation and editing of BPMN diagrams. While direct XML generation is common, it is verbose, slow, and prone to syntax errors during complex modifications. We introduce a specialized JSON-based intermediate representation designed to facilitate atomic editing operations through function calling. We evaluate our approach against direct XML manipulation using a suite of state-of-the-art models, including GPT-5.1, Claude 4.5 Sonnet, and DeepSeek V3. Results demonstrate that the JSON-based approach significantly outperforms direct XML in editing tasks, achieving higher or equivalent success rates across all evaluated models. Conformance checking evaluation confirms that generated models preserve executable semantics, with JSON achieving an average F1 score of 0.72 compared to 0.69 for XML, though frontier models like GPT-5.1 and Claude 4.5 Sonnet demonstrated superior precision with direct XML generation. Furthermore, despite requiring more input context, our approach reduces generation latency by approximately 43% and output token count by over 75%, offering a more reliable and responsive solution for interactive process modeling. Full article

Resource-constrained intelligent systems increasingly require reliable long-form reasoning capabilities under limited computational and memory budgets, particularly in edge and embedded sensing environments. However, reinforcement learning for long-horizon decision generation remains highly unstable in such low-resource settings due to severe reward sparsity and imbalanced credit assignment, which often lead to non-convergent or excessively verbose generation behavior. In this work, we propose AF-CuRL (Answer-Focused Curriculum Reinforcement Learning), a lightweight reinforcement learning framework designed to stabilize long-form generation without increasing model size or computational cost. AF-CuRL improves optimization learnability through two complementary objective-level designs: (1) answer-focused token reweighting, which concentrates policy updates on reward-critical regions of generated sequences to alleviate credit assignment imbalance, and (2) a two-phase curriculum reward schedule that prioritizes stable termination and output regularity before shifting toward correctness-oriented optimization. We evaluate AF-CuRL on a 1.5B-parameter language model under strictly constrained training settings, using mathematical reasoning tasks as a controlled and reproducible proxy for long-horizon, rule-based decision-making commonly encountered in intelligent sensing and embedded systems. Experimental results demonstrate consistent improvements in both decision accuracy and generation regularity, including higher termination reliability and reduced generation length, compared with standard sequence-level reinforcement learning baselines. These results suggest that, for resource-limited and edge-intelligent systems, structured objective design can be more effective than model scaling for achieving stable and efficient long-form reasoning, providing a practical reinforcement learning solution for intelligent systems operating under real-world constraints. Full article

Internet of Things (IoT) applications often require only minimal necessary information—such as threshold judgments, binning, or prefixes—yet they must control privacy leakage arising from multi-round and cross-entity access without exposing raw values. Existing solutions, however, frequently rely on ciphertext structures and server-side states, making it difficult to define a leakage upper bound for restricted answers in the sense of Differential Privacy (DP), or they lack unified information budgeting and k-use control. To address these challenges, this paper proposes a verifiable differential privacy partial disclosure scheme for IoT. We employ DP accounting to uniformly constrain the leakage of three types of operators: threshold, binning, and prefix. Furthermore, we design stateless k-use tokens based on Verifiable Random Functions (VRFs) and chained receipts to generate publicly verifiable compliance evidence for each response. We implemented an end-edge-cloud prototype system and evaluated its performance on two use cases: smart meter threshold alarms and industrial sensor out-of-bound detection. Experimental results demonstrate that compared with a baseline relying on server-state counting for k-use control, our stateless k-use mechanism improves throughput by approximately 25–37% under concurrency scales of 1, 8, and 16, and reduces p95 latency by an average of 15%. Meanwhile, in multi-party splicing attack experiments, the re-identification accuracy remains stable in the 0.50–0.52 range, approximating random guessing. These results validate that the proposed scheme possesses low-energy engineering feasibility and audit-friendliness while effectively suppressing splicing risks. Full article

This paper discusses a multimodal AI system applied to legal reasoning for tax law. The results given here are very general and apply to systems developed for other areas besides tax law. A central goal of this work is to gain a better understanding of the relationships between LLMs (Large Language Models) and automated theorem-proving methodologies. To do this, we suppose (1) two cases for the theorem-proving system: one where it has a countable number of total meanings for its countable number of atoms and the other is where it has an uncountable number of total meanings for its countable number of atoms, and (2) LLMs can have an uncountable number of token meanings. With this in mind, the results given in this paper use the downward and upward Löwenheim–Skolem theorems and logical model theory to contrast these two AI modalities. One modality focuses on syntactic proofs and the other focuses on logical semantics based on LLMs. Particularly, one modality uses a rule-based first-order logic theorem-proving system to perform legal reasoning. The objective of this theorem-proving system is to provide proofs as evidence of valid legal reasoning when enacted laws are applied to particular situations. These proofs are syntactic structures that can be presented in the form of narrative explanations of how the answer to the legal question was determined. The second modality uses LLMs to analyze and transform a user’s tax query so this query can be sent to a first-order logic theorem-proving system to perform its legal reasoning function. The main goal of our application of LLMs is to enhance and simplify user input and output for the theorem-proving system. Using logical model theory, we show how there can exist an equivalence between laws represented in logic of the theorem-proving system, fixed in time when the theorem-proving system was set up, and new semantics given by LLMs. These results are based on logical model theory and Löwenheim–Skolem theorems. Full article

In recent years, Transformer architectures have excelled at modeling non-local information. This makes them suitable for image deraining. However, existing methods use dense self-attention. They compute all similarities between query and key tokens. This is inefficient. In practice, this approach can lead to the neglect of the most relevant information and result in a blurring effect of irrelevant representations during the feature aggregation process. To address this issue, this paper proposes an image deraining Transformer based on sparse non-local self-attention. The core of the network consists of multiple non-local feature extraction modules, primarily comprising a sparse self-attention network and a sparse feedforward network along the channel dimension. Specifically, we implement sparse attention by selecting the most useful similarities based on Top-k approximations. Furthermore, we have developed a sparse feedforward network to achieve more accurate representations for high-quality preservation results. Extensive experiments on benchmark datasets have demonstrated the effectiveness of our proposed method. Full article

Bacterial wound infection poses a major challenge in trauma care and can lead to severe complications such as sepsis and organ failure. Therefore, rapid and accurate identification of the pathogen, along with targeted intervention, is of vital importance for improving treatment outcomes and reducing risks. However, current detection methods are still constrained by procedural complexity and long processing times. In this study, a hyperspectral imaging (HSI) acquisition system for bacterial analysis and a multi-scale dual-domain feature fusion transformer (MDF2Former) were developed for classifying wound bacteria. MDF2Former integrates three modules: a multi-scale feature enhancement and fusion module that generates tokens with multi-scale discriminative representations, a spatial–spectral dual-branch attention module that strengthens joint feature modeling, and a frequency and spatial–spectral domain encoding module that captures global and local interactions among tokens through a hierarchical stacking structure, thereby enabling more efficient feature learning. Extensive experiments on our self-constructed HSI dataset of typical wound bacteria demonstrate that MDF2Former achieved outstanding performance across five metrics: Accuracy (91.94%), Precision (92.26%), Recall (91.94%), F1-score (92.01%), and Kappa coefficient (90.73%), surpassing all comparative models. These results have verified the effectiveness of combining HSI with deep learning for bacterial identification, and have highlighted its potential in assisting in the identification of bacterial species and making personalized treatment decisions for wound infections. Full article

Iterative self-training of language models presents a promising avenue for realizing self-improving Artificial Intelligence systems; however, this process is often hindered by the fundamental challenge of “Model Collapse.” Existing research indicates that models undergo catastrophic performance degradation and diversity collapse when recursively trained on their own increasingly homogenized synthetic data. Although some data selection-based approaches attempt to mitigate this issue by enhancing diversity, they predominantly rely on static strategies, lacking a feedback mechanism capable of adapting in real-time to the dynamic evolution of the model state and data distribution. To address this limitation, we propose a dynamic data selection framework titled “DCES” (dynamic center-edge sampling). We conducted extensive experiments on iterative self-training tasks across multiple model architectures. The results demonstrate that our system significantly outperforms baselines in terms of Perplexity (PPL) and loss across various models and test sets. Simultaneously, the framework effectively mitigates the degradation of Expected Calibration Error (ECE) and entropy metrics, successfully preventing mode collapse. Our findings highlight that an adaptive system capable of intelligent data curation based on training feedback is pivotal for maintaining the dynamic balance of data distributions and achieving sustainable AI self-evolution. This work provides a systematic methodology for realizing this goal. Full article

The transition of Large Language Models (LLMs) from centralized clouds to edge environments is critical for addressing privacy concerns, latency bottlenecks, and operational costs. However, existing edge benchmarking frameworks remain tailored to discriminative Deep Learning tasks (e.g., object detection), failing to capture the multidimensional challenges of generative AI, specifically the trade-offs between token generation speed, semantic accuracy, and hardware sustainability. To address this gap, we introduce LEAF (LLM Edge Assessment Framework), a novel evaluation methodology that integrates Circular Economy principles directly into performance metrics. LEAF assesses edge deployments across five synergistic pillars: Circular Economy Score, Energy Efficiency (Joules/Token), Performance Speed (Tokens/Second), semantic accuracy (BERTScore), and End-to-End Latency. We validate LEAF through an extensive experimental analysis of five distinct hardware classes, ranging from embedded IoT devices (Raspberry Pi 4 and 5, NVIDIA Jetson Nano) to professional edge servers (NVIDIA T400) and repurposed legacy workstations (NVIDIA GTX 1050 Ti). Utilizing 4-bit quantized models via the Ollama runtime, our results reveal a counterintuitive insight: repurposed consumer hardware significantly outperforms modern purpose-built edge SoCs. The legacy GTX 1050 Ti achieved a 20× speedup over the Raspberry Pi 4 and maintained superior energy-per-task efficiency compared to low-power ARM architectures by minimizing active runtime. These findings challenge the prevailing narrative that newer silicon is essential for Edge AI, demonstrating that sustainable, high-performance inference can be achieved by extending the lifecycle of existing hardware. LEAF thus provides a blueprint for a “Green Edge” ecosystem that balances computational capability with environmental responsibility. Full article

Manufacturing companies face significant challenges in leveraging artificial intelligence for equipment management due to high infrastructure costs and limited availability of labeled data for failures. While most manufacturing AI applications focus on structured sensor data, vast amounts of unstructured textual information containing valuable maintenance knowledge remain underutilized. This study presents a practical generative AI-based framework for structured information extraction that automatically converts unstructured equipment maintenance texts into predefined semantic fields to support predictive maintenance in manufacturing environments. We adopted and evaluated three representative generative models—Bidirectional and Auto-Regressive Transformers (BART) with KoBART, Text-to-Text Transfer Transformer (T5) with pko-t5-base, and the large language model Qwen—to generate structured outputs by extracting three predefined fields: failed components, failure types, and corrective actions. The framework enables the structuring of equipment management text data from Manufacturing Execution Systems (MES) to build predictive maintenance support systems. We validated the approach using a large-scale MES dataset consisting of 29,736 equipment maintenance records from a major automotive parts manufacturer, from which curated subsets were used for model training and evaluation. Our methodology employs Generative Pre-trained Transformer 4 (GPT-4) for initial dataset construction, followed by domain expert validation to ensure data quality. The trained models achieved promising performance when evaluated using extraction-aligned metrics, including exact match (EM) and token-level precision, recall, and F1-score, which directly assess field-level extraction correctness. ROUGE scores are additionally reported as a supplementary indicator of lexical overlap. Among the evaluated models, Qwen consistently outperformed BART and T5 across all extracted fields. The structured outputs are further processed through domain-specific dictionaries and regular expressions to create a comprehensive analytical database supporting predictive maintenance strategies. We implemented a web-based analytics platform enabling time-series analysis, correlation analysis, frequency analysis, and anomaly detection for equipment maintenance optimization. The proposed system converts tacit knowledge embedded in maintenance texts into explicit, actionable insights without requiring additional sensor installations or infrastructure investments. This research contributes to the manufacturing AI field by demonstrating a comprehensive application of generative language models to equipment maintenance text analysis, providing a cost-effective approach for digital transformation in manufacturing environments. The framework’s scalability and cloud-based deployment model present significant opportunities for widespread adoption in the manufacturing sector, supporting the transition from reactive to predictive maintenance strategies. Full article