Retrieval-Augmented Generation (RAG) systems navigate a fundamental trade-off between the computational simplicity of flat vector indices and the structural richness of graph representations. Standard dense retrieval collapses document discourse into isolated embeddings, causing semantic fragmentation and context loss....
Alex Martins Almeida Neves· Zenodo (CERN European Organi...· 0 citations
This dataset contains photorealistic, high-resolution AI-generated facial portraits created by the authors to support research in synthetic media forensics, deepfake detection, and generative source attribution. Existing synthetic face benchmarks focus heavily on legacy Generative Adversarial Networks (GANs) or early l...
Generative AI can augment human productivity and cognitive capability. Existing discussions of AI dependence have addressed cognitive offloading, overreliance, skill degradation, delegation of cognitive agency, and the social diffusion of AI use. This paper proposes a complementary pathway: rational AI dependence throu...
kurato· Zenodo (CERN European Organi...· 0 citations
Abstract Background Accurate modelling of allele frequency trajectories requires incorporation of both genetic mechanisms such as selection and mutation, as well as realistic population demography. Accounting for a non-constant demography within a Wright–Fisher diffusion framework induces a time-inhomogenous drift coef...
Jaromir Sant, Paul A. Jenkins, Jere Koskela et al.· BMC Bioinformatics· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work systematically categorizes representative methods across a wide range of downstream tasks, including image classification, dense visual prediction, instance-level perception, and annotation-scarce scenarios, and proposes a three-tier progressive framework that organizes existing works from three perspectives.
Abstract. We propose a reduced axisymmetric tornado model in which the vertical transport of momentum is nonlocal, governed by a fractional derivative whose order is not a free parameter but is fixed by the observed coherent substructure of the vortex through the hierarchical cascade relation α = 2−2/(N+1). Building on...
The nearly edge-on starburst galaxy NGC 253 has been observed to exhibit extended halo emission in multiple bands, making it an ideal laboratory for studying the transfer of matter from the disk to the halo. We aim to determine how the cosmic-ray electrons (CREs) flow from the disk to the halo and understand what drive...
Shengtao Wang, George Heald, Stefan W. Duchesne et al.· Astronomy and Astrophysics· 0 citations
Mutualisms are key for structuring ecological communities, but they are sensitive to environmental change and fluctuations in population size. Consequently, how mutualisms achieve stability remains an open question in ecological theory. Motivated by previous results in competitive and predator–prey interactions, and by...
M. Bongestab, David Pinto-Ramos, Ricardo Martínez‐García· PLOS Ecosystems· 0 citations
This study captures in real time, the VOC-induced rapid signalling in receiver plants and its ensuing effect on growth, an important advance in deciphering how emitter-specific volatile cues are integrated into plant responses, guiding rational selection of beneficial companion plants for improved yield and nutritional...
Mervin Chun-Yi Ang, Jolly Madathiparambil Saju, Sayyid Mohaideen et al.· Plant, Cell and Environment· 0 citations
A sequentially gated theranostic nanoplatform based on a silicon-rhodamine (Si-Rhod) photocage that integrates endogenous reactive oxygen species sensitivity with externally applied red light activation to achieve spatiotemporally programmable chemotherapy is reported.
Naga Mallika Reddy, Amar Singh, Soumyadeep Poddar et al.· Journal of materials chemist...· 0 citations
Electrochemical reaction rates are controlled by both interfacial charge-transfer kinetics and reactive-ion transport. In charged porous media, fixed charges enrich reactive counterions near pore walls, enabling transport beyond the classical diffusion limit via surface conduction (SC). Under overlimiting conditions, c...
Arjun V. Yennemadi, Junghyun Yoon, Martin Z. Bazant· 0 citations
A new computational approach reveals how cells communicate to assemble into tissues. This could accelerate stem cell engineering, improve disease models, and inform therapies.
Learn how Google DeepMind and GovTech Singapore uses lightweight on-device Gemma models to intercept conversational scams in private chats with zero cloud data egress.
Microsoft Research Blog· microsoft.comOct 12, 2026
StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics. The post Simulating student learning for better AI tutoring appeared first on Microsoft Research.