3031 papers
This paper proposes a weakly supervised framework for transaction-graph anomaly detection that leverages optimal transport to align hierarchical rule semantics, derived from decision trees, with continuous transaction representations, thereby generating high-quality pseudo-labels and outperforming existing baselines on benchmark datasets.
This paper argues that current public regulatory records for AI medical devices are inadequate for lifecycle oversight due to a lack of structured, queryable data on software changes and AI-specific attributes, and proposes a new data specification to address these critical deficits across major global jurisdictions.
This paper proposes a multimodal unsupervised anomaly detection framework that fuses textual semantics from OCR and financial logical features to effectively identify structural irregularities and fraud in semi-structured invoices, outperforming single-modality baselines on real-world data.
This paper introduces CLASP, an adaptive constrained reinforcement learning scheduler that enforces explicit latency constraints via a PID-controlled Lagrangian price and price-conditioned policy, thereby maintaining stable energy efficiency and violation targets across varying workloads where traditional fixed-weight reward approaches fail.
This paper proposes and evaluates a multi-layer adaptive Zero Trust framework that utilizes a context-aware trust scoring mechanism and rigorous mathematical modeling to demonstrate a significant reduction in attack success rates and lateral movement within cloud environments compared to traditional perimeter-based security.
RadioAI presents a reproducible, physics-grounded multi-modal deep learning system that integrates specialized architectures for chest CT, breast ultrasound, and brain MRI classification with an automatic modality-routing Gatekeeper and interpretable visualization tools, validated through controlled ablation studies demonstrating the empirical necessity of physics-informed design components.
The paper introduces QuantumRAG, a universal, low-latency retrieval system that maps text embeddings to complex quantum states in Hilbert space to achieve BERT-level accuracy with 3.6× faster inference on standard CPU hardware, while remaining agnostic to data modalities like text, math, and video.
This paper introduces the MAANG-AI-450 dataset to analyze the educational backgrounds and citation impacts of AI researchers at major tech companies, revealing that while PhD training is concentrated at elite institutions, citation success is highly skewed and often driven by current work environments rather than doctoral pedigree alone.
This paper introduces Deep Geometric and Quantum Geometric Networks (DGN/QGN), a neuro-symbolic architecture that achieves competitive gesture recognition by extracting explicit, interpretable differential-geometric features from motion and mapping them directly to quantum circuit parameters, demonstrating that current performance limitations stem from hardware constraints rather than algorithmic flaws.
This study integrates bibliometric analysis with an AI paradigm perspective to map the structural evolution of pedestrian detection research, revealing that the field's transition from handcrafted features to deep learning and transformers occurs through gradual knowledge integration rather than abrupt shifts, thereby offering a novel theoretical framework for understanding AI development.