ArXiv

Authors Brockers VC, Ventzke RD, Neuhaus V, Hidalgo-Ogalde B, Priesemann V Journal Arxiv Citation arXiv:2605.23645 Abstract In the context of artificial neural networks, subliminal learning refers to the transfer of task-relevant knowledge or unintended biases from teacher to student models through distillation on task-unrelated input output pairs. Prior explanations tie

Authors Golovin A, Mohr SB, Gottwald AI, Hvid U, Trivedi S, Neto JP, Schneider AC, Priesemann V Journal Arxiv Citation arXiv:2605.15956 Abstract Here we present a massive longitudinal dataset of public Telegram content, comprising over 5.9 billion messages dating from 2015 to 2025, collected from 712 thousand channels and groups,

Authors Arenas-Velilla S, Munk A, Rodríguez LA Journal Arxiv Citation arXiv:2604.12940 Abstract Recently, the statistical properties of empirical Entropic Optimal Transport (EOT) have attracted great interest, as this quantity has been shown to be useful for complex data analysis, among other reasons due to its computational efficiency. In several applications,

Authors Fischer J, Jurado A, Betz T Journal Arxiv Citation arXiv:2604.14918 Abstract Recent advances in the field of active soft matter promise a lot. Both, experimental advances and theoretical understanding point towards new material classes in reach, for example self-healing materials that might switch their properties from elastic to solid

Authors Teuber C, Archit A, Boothe T, Ditte P, Rink J, Pape C Journal Arxiv Citation arXiv:2603.19802. Abstract Deep learning underlies most modern approaches and tools in computer vision, including biomedical imaging. However, for interactive semantic segmentation (often called pixel classification in this context) and interactive object-level classification (object classification),

Authors Archit A, Pape C Journal Arxiv Citation arXiv:2603.17845. Abstract Cell segmentation is a fundamental task in microscopy image analysis. Several foundation models for cell segmentation have been introduced, virtually all of them are extensions of Segment Anything Model (SAM), improving it for microscopy data. Recently, SAM2 and SAM3 have

Authors Dvoretskii S, Archit A, Pape C, Moore J, Nolden M Journal Arxiv Citation arXiv:2512.15820. Abstract Modern bioimage analysis approaches are data hungry, making it necessary for researchers to scavenge data beyond those collected within their (bio)imaging facilities. In addition to scale, bioimaging datasets must be accompanied with suitable, high-quality

Authors Müller L, Sartori F, Dehning J, Eggl MF, Priesemann V Journal Arxiv Citation arXiv:2512.11454. Abstract Mitigation measures are essential for controlling the spread of infectious diseases during pandemics and epidemics, but they impose considerable societal, individual, and economic costs. We developed a general optimization framework to balance costs related

Authors Blümel M, Schneider AC, Neuhaus V, Ehrlich DA, Graetz M, Wibral M, Makkeh A, Priesemann V Journal Arxiv Citation arXiv:2511.02584 Abstract Associative memory, traditionally modeled by Hopfield networks, enables the retrieval of previously stored patterns from partial or noisy cues. Yet, the local computational principles which are required to

Authors Requadt R, Li H, Munk A Journal Arxiv Citation arXiv:2510.05645. Abstract We investigate the asymptotic behavior of parametric Bayes estimators under a broad class of loss functions that extend beyond the classical translation-invariant setting. To this end, we develop a unified theoretical framework for loss functions exhibiting locally polynomial