2608 papers
This paper proposes a heterogeneity-aware evaluation framework for federated learning that conditions privacy–utility–fairness conclusions on measured realized client distributions rather than relying solely on nominal Dirichlet parameters, demonstrating that equal concentration values yield significant variation in actual data structure and performance.
This paper presents WVAB, an offline-first assistive vision system that utilizes temporal risk-gated multimodal guidance to significantly reduce unstable focus switching and information overload for blind users, while demonstrating a measurable trade-off between guidance stability and responsiveness to rapidly changing hazards alongside verified edge-streaming security.
This paper proposes an identity-aware permissioned blockchain architecture that secures connected vehicle communications by separating real-time message propagation from ledger ordering, thereby achieving high throughput and low-latency finality while effectively mitigating identity spoofing, replay attacks, and validator compromises.
This paper proposes a permissioned-blockchain architecture for connected vehicles that ensures forensic integrity and secure incident recovery by storing detailed evidence in a local tamper-evident journal while committing compact cryptographic checkpoints to a distributed ledger, thereby achieving high-performance verification and significant data volume reduction without compromising privacy or safety.
This paper proposes a hybrid authentication architecture for permissioned vehicular blockchains that combines post-quantum key establishment with session-bound message authentication codes and a trust-aware admission policy to drastically reduce communication overhead and latency while maintaining robust security against insider attacks across vehicle lifetimes.
This paper critically evaluates a TimeGAN framework for generating synthetic Nigerian Naira/US Dollar scenarios under data scarcity, revealing that the model's outputs are dominated by classical baselines in distributional fidelity and exhibit extreme seed-dependent instability, thereby arguing that GAN-based scenario generation requires specific architectural priors and rigorous multi-seed reporting before it can responsibly complement conventional stress-testing tools.
This paper proposes the Unified Artificial Intelligence Framework for Industry 5.0 (UAI-5), a multi-layered architecture integrating Multi-Agent Systems with a hybrid ADMM-LASSO optimization strategy to overcome fragmentation and scalability issues, thereby enabling robust, real-time, and sustainable decision-making across diverse industrial and societal domains.
SFCT-Net is a robust medical image segmentation framework that redefines the task as spatial-frequency signal reconstruction by integrating synchronized convolutional-attention streams with a Wavelet-Driven Reconstruction Module to effectively suppress noise and preserve high-frequency anatomical boundaries across diverse imaging modalities.
This paper proposes a robust industrial defect detection framework that integrates Sparse Principal Component Analysis and a sparsified Extreme Learning Machine, with hyperparameters optimized by the Capuchin Search Algorithm, to effectively mitigate sensor noise and adversarial perturbations while maintaining high classification accuracy.
This paper proposes an Ordinal Multi-learner Ensemble (OME) augmented with a Hidden Markov Model generator to effectively predict imbalanced, ordinal post-therapy occupational performance scores in children with cerebral palsy, demonstrating superior accuracy and robustness compared to existing baselines and SMOTE.