AI Paper Atlas

Research map / Graph learning / GNN architectures

Differentiable Lifting for Topological Neural Networks

Jorge Luiz Franco, Gabriel Duarte, Alexander V Nikitin, Moacir A Ponti, Diego Mesquita, Amauri H Souza

ICLR 2026 · Graph learning › GNN architectures · approach: Physics-inspired GNNs · 1 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
replace cycle lifting with differentiable liftingCWN · NCI1accuracy: 76.93 -> 79.81Improved
replace cycle lifting with differentiable liftingCWN · MUTAGaccuracy: 66.67 -> 85.96Improved
replace cycle lifting with differentiable liftingCWN · ZINCMAE: 0.46 -> 0.17Improved
replace cycle lifting with differentiable liftingCXN · NCI1accuracy: 72.02 -> 82.08Improved
replace cycle lifting with differentiable liftingCXN · Proteinsaccuracy: 70.83 -> 69.94Hurt
replace cycle lifting with differentiable liftingCIN · MUTAGaccuracy: 66.96 -> 88.72Improved
replace k-hop lifting with differentiable liftingUniGCN2 · MOLHIVaccuracy: 50.72 -> 69.32Improved
replace k-NN lifting with differentiable liftingUniGIN · NCI1accuracy: 72.83 -> 64.88Hurt
replace GIN backbone with GPS in dLift (ablation)CWN · ZINCMAE: 0.46 -> 0.17Improved
replace GIN backbone with GPS in dLift (ablation)CWN · MOLHIVaccuracy: 75.37 -> 64.31Hurt
replace cycle lifting with differentiable liftingCWN · Node classification averageaccuracy: 73.52 -> 77.74Improved
replace cycle lifting with differentiable liftingTopoTune · Node classification averageaccuracy: 70.91 -> 75.84Improved

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