2890 papers
This paper presents a resource-optimized and low-power SHA-256 hardware accelerator designed for edge devices, which utilizes 2X loop unrolling and clock gating to achieve a throughput of 1.47 Mhps and an energy efficiency of 13.37 Mhps/W on an FPGA platform.
This paper proposes a self-evolving agentic reinforcement meta-learning framework that integrates goal evolution with policy learning to autonomously adapt cloud scheduling objectives in response to dynamic workloads, demonstrating superior performance in latency, cost, and resource utilization compared to traditional scheduling methods.
This paper proposes SecureChain-ID, a blockchain-powered identity management system that leverages elliptic curve cryptography and zero-knowledge proofs to provide scalable, secure, and privacy-preserving authentication for resource-constrained IoT devices while eliminating single points of failure through decentralized governance.
This paper proposes a data-driven framework that integrates guided BERTopic, a dynamic domain-adapted sentiment lexicon, and the Kano model to objectively analyze online reviews and prioritize product improvements, demonstrated through an evaluation of 8,970 new energy vehicle reviews.
This paper proposes a multi-agent parameter-sharing Proximal Policy Optimization (PPO) framework for dynamic mode selection and power allocation in clustered vehicular networks, which significantly outperforms traditional baselines by improving reliability, reducing latency, and lowering transmit power through adaptive learning of local network conditions.
This paper introduces the Symbiotic Convolution Block (SCB), a lightweight, non-attentive module that achieves implicit feature recalibration and diversification through fused convolutional maps, offering a computationally efficient alternative to attention mechanisms for image classification tasks.
This study proposes a hybrid framework that integrates a fine-tuned DeBERTa-v3 model, which effectively captures neutral sentiment in financial news, with technical indicators to significantly improve the directional prediction accuracy of the NIFTY 50 index compared to price-only baselines.
This paper proposes GeoRel-CLIP, a geometry-regularized relational distillation framework that enhances CLIP's zero-shot recognition by jointly optimizing global geometric distribution and local relational structure to mitigate modality gaps and representation collapse during post-pretraining.
This study proposes an explainable, label-efficient machine learning framework using CatBoost and semi-supervised learning to accurately recognize contact configurations in tendon-driven continuum robots for minimally invasive surgery, achieving high performance with limited labeled data while identifying normal force as a critical feature for safer control strategies.
This paper presents an AI-driven profiling model within a cyber range environment that continuously evaluates learner performance to generate automated, personalized feedback and adaptive learning paths, thereby addressing key challenges in scalability and personalization for cybersecurity education.