On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,291 papers · 148 categories

Trend · papers per month

count % of corpus

6 results for “G-flow”

Neck Detection for Two-Convex Hypersurfaces Embedded in Euclidean Space undergoing Brendle-Huisken G-Flowmath.DG

This paper detects necks in two-convex hypersurfaces under a specific flow.

problem Detecting necks in two-convex hypersurfaces under a new flow.

method Adjusting gradient estimates and following Huisken-Sinestrari's surgery algorithm.

result Classifies two-convex surfaces undergoing the G-flow.

neck detection two-convex hypersurfaces G-flow Brendle-Huisken flow surgery algorithm

2018-03-27abs ↗pdf ↗

CoSE: Compositional Stroke Embeddingscs.LG

CoSE models complex drawings by treating strokes as a collection that can be composed.

problem Modeling complex free-form structures like diagrams.

method Generative model using autoencoder to project strokes into a fixed latent space, relational model operates in latent space.

result Model captures both individual strokes and their compositional structure.

stroke-based drawing autoencoder latent space compositional structure diagrams

2020-06-17abs ↗pdf ↗

Non-uniqueness in Mean Curvature Flow: Non-canonical solutions via the parabolic Allen--Cahnmath.AP

New non-canonical flows found via parabolic Allen-Cahn equations.

problem Existence of non-canonical mean curvature flows inside fattening regions.

method Construction of non-canonical flows as limits of parabolic ε-Allen-Cahn solutions.

result First examples of non-outermost, non-canonical integral Brakke motions.

mean curvature flow non-uniqueness fattening parabolic Allen-Cahn integral Brakke motions

2025-10-08abs ↗pdf ↗

Dynamic Node Embeddings from Edge Streamscs.LG

New method learns dynamic node embeddings from temporal networks.

problem Temporal networks' temporal information is often ignored or approximated.

method Continuous-time dynamic network embeddings (CTDNEs) using temporal walks.

result CTDNEs achieve an average gain of 11.9% in AUC across all methods and graphs.

networks temporal embeddings dynamic walks

2019-04-12abs ↗pdf ↗

Causal Holography in Application to the Inverse Scattering Problemsmath.GT

Causal holography reconstructs manifold properties from scattering data.

problem Reconstructing manifold properties from scattering data.

method Introducing metrics of gradient type and using scattering maps.

result Reconstruction of manifold's homology, Gromov simplicial semi-norm, and fundamental group.

causal holography inverse scattering manifold reconstruction scattering maps geodesic flow

2017-03-26abs ↗pdf ↗

Holography of geodesic flows, harmonizing metrics, and billiards' dynamicsmath.DS

Study geodesic flows, billiards, and metrics on manifolds.

problem Inverse scattering and billiard dynamics on Riemannian manifolds.

method Lyapunov function, harmonizing metrics, isoperimetric inequalities, Santaló-Chernov formulas.

result Holography theorems and formulas for geodesic and billiard dynamics.

geodesic flows billiards Riemannian manifolds harmonizing metrics isoperimetric inequalities

2020-03-23abs ↗pdf ↗