(1) Background: Although numerous artificial intelligence (AI)-based air pollution prediction models have been proposed, research that links key pollution drivers, such as regional industrial facilities, to actionable policy recommendations is required. (2) Methods: This study employs the radial basis function (RBF) and spatial lag features to capture spatial interactions among regions, utilizing a transformer model for analysis. The model was trained on air quality and industrial data from South Korea (2010–2022) and Japan (2017–2020). (3) Results: The transformer model achieved a mean squared error of 0.045 for the Korean dataset and 0.166 for the Japanese dataset, outperforming benchmark models, including Support Vector Regression, neural networks, and the AutoRegressive Integrated Moving Average model. (4) Conclusions: By capturing complex spatial dynamics, the proposed model provides valuable insights that can assist policymakers in developing effective, data-driven strategies for air pollution reduction at the national and regional levels, thereby supporting the broader goals of sustainability through informed, equitable environmental interventions.

Split learning is a collaborative learning paradigm designed to reduce client-side computational load while providing inherent privacy advantages by splitting up a model among the client and the server. However, it faces a critical vulnerability: an honest-but-curious server can reconstruct the original client data from the intermediate representation of the client model, called smashed data, which is transmitted to the server. Prior defenses typically inject random noise into smashed data to reduce privacy leakage. Although noise-based defenses enhance privacy, the presence of noise hinders server model training, resulting in reduced overall model performance. To effectively defend privacy leakage against original data reconstruction attacks targeting smashed data while minimizing model performance degradation, we propose the Personal Embedding Transformation and Contrastive Representation Separation module. The Personal Embedding Transformation module randomly applies a set of differentiable transformation functions to the embedding dimensions within the client model. These transformations diversify and reshape the embedding space in a client-specific manner. Contrastive Representation Separation reorganizes the transformed embeddings through contrastive learning, clustering representations of the same class while pushing apart those of different classes. The combination of these mechanisms produces client-specific embedding spaces that distort sensitive information while retaining class-level discriminative features for downstream tasks. The proposed mechanisms enhance privacy guarantees in the split learning setting while mitigating the severe downstream model utility degradation typically associated with noise-based defenses. Experiments on large-scale text classification demonstrate that our framework achieves strong protection against reconstruction attacks while incurring negligible degradation in model utility and computational efficiency. Specifically, our method introduces only a ∼2.56% increase in client-side inference latency, confirming its practical feasibility. In particular, the proposed method reduces reconstruction similarity by 34% in cosine score and lowers text-level overlap metrics such as BLEU and ROUGE by more than 33% compared to noise-based defense, indicating that adversarially reconstructed texts exhibit significantly reduced semantic similarity to the original inputs.

Phishing attacks via desktops, smartphones and internet of things devices are becoming increasingly sophisticated, posing critical security challenges for digital infrastructures. Defending against these attacks requires AI-based detection models that maintain high accuracy, since false positives or negatives can lead to severe breaches, while remaining lightweight enough to run on resource-constrained client devices. Split Learning (SL) meets these requirements by having clients compute only initial model layers locally and transmit intermediate activations (“smashed data”) to a server for the remaining inference, avoiding direct sharing of raw inputs. However, prior work in the image domain has shown that smashed data can leak original content, suggesting that SL may not be safe for user privacy. Therefore, it is essential to investigate whether these privacy risks also extend to language-model–based SL systems, which have fundamentally different neural network architectures, including attention mechanism. This paper introduces the Semantic Information Reconstruction Attack (SIRA), a novel framework designed to infer sensitive semantic elements directly from smashed data by leveraging the generative capabilities of large language models. In experiments on real-world phishing datasets, SIRA outperforms conventional reconstruction attacks in accurately inferring private webpage information. These findings reveal a potential privacy vulnerability in SL-based language models for security applications and motivate the development of targeted defense strategies.