863 papers
This paper introduces the Oversight-to-Control (O2C) framework, a socio-technical model comprising six functions and supporting conditions, to diagnose why human oversight often fails to achieve effective control in high-risk AI systems by analyzing regulatory instruments and real-world cases where interpretation and timely intervention frequently break down.
This paper proposes AJIT, a high-fidelity classical–quantum hybrid framework that ensures quantum-resilient secure sharing and multiparty authentication of DICOM medical images by combining DTCWT-based compression, quasi-periodic unitary scrambling, and topological hypergraph verification to achieve lossless reconstruction and robust differential resistance.
This study introduces the Cross-Market Retrieval-Augmented Lag-Llama (CM-RAF-Lag-Llama) framework, demonstrating that integrating historical retrieval with a pre-trained Lag-Llama model significantly reduces forecasting errors across diverse equity markets by leveraging analogous historical windows to correct predictions, particularly for volatility and volume-related targets.
The paper proposes CFE-UNet, a novel semantic segmentation framework for remote sensing that integrates a reconstructed encoder with a Cross-Feature Exchange decoder to effectively combine global context and local details, achieving state-of-the-art performance on complex, heterogeneous urban and natural scenes.
This paper demonstrates that in restricted multimodal oncology settings with limited and heterogeneous data, auditable evidence-role organization—assigning specific, compact roles to molecular, phenotypic, and clinical inputs—is a more critical determinant of cross-domain transfer performance than simply increasing representation scale.
This study proposes a GraphRAG-based methodology for automated meta-analysis of Transcranial Magnetic Stimulation (TMS) literature, which utilizes title-based chunking, paper-level knowledge graph construction, and hybrid retrieval to achieve high-accuracy parameter extraction and significantly reduce the time required for evidence synthesis.
StructFix is a structure-aware automated program repair framework that enhances masked language models by integrating Code Property Graphs to better capture control and data dependencies, thereby improving repair effectiveness and cross-language robustness compared to existing token-sequence-based approaches.
The paper proposes FedVIB–AGP, a post-aggregation repair framework for federated learning that combines Variational Information Bottleneck regularization during training with Activation-Gap Pruning during repair to effectively suppress distributed backdoor attacks while maintaining high clean-task accuracy.
This paper demonstrates that while systematically varying prompt engineering elements like label descriptions, instructional nudges, and few-shot examples can significantly improve LLM classification accuracy for social science texts, performance gains are often marginal beyond minimal context increases, can sometimes decline with excessive context, and vary substantially across models and tasks, necessitating individual validation rather than reliance on general rules.
This paper proposes a practical, multi-layer deep learning framework utilizing CNNs and LSTMs to detect advanced cybersecurity threats by addressing critical engineering challenges like data imbalance and providing a guide for scalable, real-time implementation in Security Operations Centers.