Statistically Meaningful Geometry and Gauge Symmetry Breaking: A Geometric Foundation for Scientific Discovery and Intelligence Emergence

This arXiv feed (multiple 2026-07-08 submissions) covers a range of machine-learning and applied-ML research: a theoretical framework (Statistically Meaningful Geometry) for distinguishing genuine discovery vs. spurious interpolation in overparameterized models; Design-CP, a multi‑GPU context-parallel sharding strategy that enables end-to-end all-atom design of very large symmetric protein nanoparticles; GAIA, a geometry-aware denoiser for UWB infrastructure‑anchored work‑zone reconstruction; the "Granularity Paradox" in time-series forecasting; Exogenous Dropout, a simple training method that