Chun-houh Chen, Shun-Chuan Chang, Chiun-How Kao +5 more
Continuous data has a rich toolbox of plots. Categorical data, the kind full of yes/no, blood-type, or species labels, has far less, and existing options tend to shrink everything into an abstract low-dimensional scatter that loses touch with the actual table you started from. cGAP tries to keep the table.
It runs Homogeneity Analysis to place both subjects and category levels in a 3D space, then paints that space onto red-green-blue values so rows with similar profiles end up similarly colored. Three linked views work together: a colored heatmap of the raw matrix plus proximity matrices for subjects and variables, with seriation reordering rows and columns to surface clusters and outliers. The authors also prove properties about how faithfully the geometry survives the trip to color, demonstrating it on datasets from mushrooms to mammalian teeth to gene families.
This is an exploratory visualization tool, so the paper's figures matter more than usual for judging it.
High-dimensional categorical data arise in genetics, biomedicine, and the social sciences, yet visualization tools for such data remain far less developed than those for continuous variables. Existing methods either scale poorly, rely heavily on low-dimensional displays detached from the original data matrix, or prioritize predictive accuracy over interpretability. To address this gap, we introduce categorical Generalized Association Plots (cGAP), a visualization framework…
DAPGNet: Dynamic Adaptive Physics-Guided Graph Diffusion Network for Hyperspectral Image Classification
arXiv (cs.AI) · July 16, 2026NIFA: Nonlinear IMC enhanced FPGA for efficient ML inference
arXiv (cs.LG) · July 16, 2026Learning in Infinitesimal Non-Compositional Sketches
arXiv (cs.AI) · July 16, 2026Long-Context Fine-Tuning with Limited VRAM