Answers are grounded in this paper and cite their sources.
Geometric and empirical demonstration that negative-eigenvalue geodesic curvature g3 on connectivity-guaranteed MST+kNN hidden-activation graphs cleanly separates neural network generalization (monotonic manifold flattening, -43% to -73%) from random-label memorization (active manifold tangling, +11% to +75%) across ResNet-CNNs and 5-seed MLPs (>7 sigma separation, robust across neighborhood scales k) where linear covariance rank and Shannon effective rank fail or invert by up to 2.8x. Direct audits of memorized training clouds uncover intermediate-layer training tangling and a label-free readout Train-Test curvature gap (Delta g3 > 0.54), alongside two-stage asymmetric flattening in grokking.
No takes yet. Share an insight, caveat, or question.
Shoji Fujita (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: