arXiv:2602.19528v1 Announce Type: cross Abstract: Crash classification models in transportation safety are typically evaluated using accuracy, F1, or AUC, metrics that cannot reveal whether a model is silently overfitting. We introduce a spectral diagnostic framework grounded in Random Matrix Theory (RMT) and Heavy-Tailed Self-Regularization (HTSR) that spans the ML taxonomy: weight matrices for BERT/ALBERT/Qwen2.5, out-of-fold increment matrices for XGBoost/Random Forest, empirical Hessians for Logistic Regression, induced affinity matrices for Decision Trees, and Graph Laplacians for KNN. Evaluating nine […]

Ver mais

With the rise of generative AI, creative ownership has been a hot topic worldwide. As AI tools like ChatGPT, Gemini, and others are becoming much more capable of generating images, text, and music independently, legal questions about copyright infringement and authorship are intensifying. Who truly owns content generated by AI, and can it ever qualify for legal protection? Well, the U.S. Supreme Court may have just answered that question. Thaler’s AI artwork sparks landmark copyright debate As reported […]

Ver mais

Utilizing the third-party library, the safe framework integration system resolves the serious problem of dependency risk and license violation in software development. This means that we are trying to place security mechanisms into the framework that works to defend against vulnerabilities that are introduced by third-party components. It also includes Open-Source Software (OSS) governance automation to monitor and enforce compliance and license obligations, and control legal and operational risks. Merging secure integration practices with automated governance allows organizations […]

Ver mais

arXiv:2602.23559v1 Announce Type: new Abstract: We present the first study of cross-sensor view synthesis across different modalities. We examine a practical, fundamental, yet widely overlooked problem: getting aligned RGB-X data, where most RGB-X prior work assumes such pairs exist and focuses on modality fusion, but it empirically requires huge engineering effort in calibration. We propose a match-densify-consolidate method. First, we perform RGB-X image matching followed by guided point densification. Using the proposed confidence-aware densification and self-matching filtering, we […]

Ver mais

Constrained Object Hierarchies (COH) offer a unified theoretical foundation for artificial general intelligence (AGI), rooted in neuroscience principles and developed with full mathematical rigor. This paper presents the complete formalization of COH theory, showing how intelligence emerges from hierarchically composed structures governed by adaptive optimization constraints. We introduce precise definitions, establish core properties of soundness and completeness, and situate COH within established mathematical frameworks including category theory, dynamical systems, and information theory. Building on this foundation, we prove […]

Ver mais

arXiv:2602.05852v1 Announce Type: cross Abstract: Community detection in networks is a fundamental problem in machine learning and statistical inference, with applications in social networks, biological systems, and communication networks. The stochastic block model (SBM) serves as a canonical framework for studying community structure, and exact recovery, identifying the true communities with high probability, is a central theoretical question. While classical results characterize the phase transition for exact recovery based solely on graph connectivity, many real-world networks contain additional […]

Ver mais