Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs

Collection of new 2026-stat/ML arXiv submissions covering unsupervised domain adaptation with kernel GLMs and pseudo-labeling, fast spectral embeddings for evolving graphs, near-equivalent Q-learning policies for dynamic treatment regimes, memorization effects in learned generative priors for geophysical inverse problems (including diffusion models), model selection and estimation for multi-dimensional GMMs, efficient hyperparameter selection for finite-context models, an ensemble approach to explainable clustering, a minimax generalized cross-entropy loss for robust classification, a survey/"