# A Neural Tension Operator for Curve Subdivision across Constant Curvature Geometries

Source: [A Neural Tension Operator for Curve Subdivision across Constant Curvature Geometries](<https://www.sota2.com/research/paper/69cca6700636e5bf4afcba6e>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-02T05:00:04.464Z

Authors: Hassan Ugail, Newton Howard

Published: 2026-03-30T19:16:25.000Z

## Abstract

Interpolatory subdivision schemes generate smooth curves from piecewise-linear control polygons by repeatedly inserting new vertices. Classical schemes rely on a single global tension parameter and typically require separate formulations in Euclidean, spherical, and hyperbolic geometries. We introduce a shared learned tension predictor that replaces the global parameter with per-edge insertion angles predicted by a single 140K-parameter network. The network takes local intrinsic features and a trainable geometry embedding as input, and the predicted angles drive geometry-specific insertion operators across all three spaces without architectural modification. A constrained sigmoid output head enforces a structural safety bound, guaranteeing that every inserted vertex lies within a valid angular range for any finite weight configuration. Three theoretical results accompany the method: a structural guarantee of tangent-safe insertions; a heuristic motivation for per-edge adaptivity; and a conditional convergence certificate for continuously differentiable limit curves, subject to an explicit Lipschitz constraint verified post hoc. On 240 held-out validation curves, the learned predictor occupies a distinct position on the fidelity--smoothness Pareto frontier, achieving markedly lower bending energy and angular roughness than all fixed-tension and manifold-lift baselines. Riemannian manifold lifts retain a pointwise-fidelity advantage, which this study quantifies directly. On the out-of-distribution ISS orbital ground-track example, bending energy falls by 41% and angular roughness by 68% with only a modest increase in Hausdorff distance, suggesting that the predictor generalises beyond its synthetic training distribution.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Ground-track evaluation on ISS ground-track (test)](<https://www.sota2.com/research/sota/ground-track-evaluation-on-iss-ground-track-test.md>): Ground-track evaluation; ISS ground-track (test); Hausdorff Distance: 0.00699.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.28937>)

[PDF](<https://arxiv.org/pdf/2603.28937.pdf>)
