AI Paper Atlas

Research map / Graph learning / GNN architectures

An Algebraic Geometry Approach to Viewing Graph Solvability

Federica Arrigoni, Kathlén Kohn, Andrea Fusiello, Tomáš Pajdla

TPAMI 2026 · Graph learning › GNN architectures · approach: Substructure-aware GNN expressivity · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use node-based Jacobian formulation instead of edge-based solvability matrixfinite solvability test · Real SfM graphs, Alamo to Piccadillytotal time (s): 25889.20 -> 2361.20 (Piccadilly)Improved
use node-based Jacobian formulation instead of edge-based solvability matrixfinite solvability test · Real SfM graphs, Skansen Kronan to Roman Forumtotal time (s): 913.57 -> 158.16 (Roman Forum)Improved
use node-based Jacobian formulation on small graphsfinite solvability test · Real SfM graphs, small (Gustav Vasa)total time (s): 0.10 -> 0.37No clear effect
use new Jacobian-based test instead of Arrigoni et al.finite solvability test · Random graphs, 20 nodes, 8000 samplesagreement of outputs: same output on all 8000 samplesNo clear effect
use new Jacobian-based test instead of Arrigoni et al.finite solvability test · Real SfM graphsnumber of components: 4 -> 4 (Tower of London)No clear effect
lower edge density of random graphsfinite solvability · Random graphs, 20 nodes# finite solvable out of 1000: 1000 -> 253 (60% to 20% density)Hurt
lower edge density of random graphsfinite solvability · Random graphs, 20 nodesnumber of components: [1, 1] -> [10, 25] (60% to 5% density)Hurt
use node-based unknowns (12n) instead of edge-based (16m)matrix size · Formulation comparison# columns: 16m -> 12nImproved
use direct formulation rows 10m+n+15 instead of 23m-11n+15matrix size · Formulation comparison# rows: 23m - 11n + 15 -> 10m + n + 15Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy