AI Paper Atlas

Research map / Graph learning / GNN architectures

FlowSymm: Physics–Aware, Symmetry–Preserving Graph Attention for Network Flow Completion

Ege Demirci, Francesco Bullo, Ananthram Swami, Ambuj Singh

ICLR 2026 · Graph learning › GNN architectures · approach: Expressive equivariant message passing · 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
add edge-wise attention over group-action basis instead of uniform random actions with one global weight (ablation)FlowSymm (k=256) · Traffic, Power, BikeRMSE: +5% (Traffic), +11.5% (Power), +6.8% (Bike)Improved
add group-action basis modeling to attention encoder (ablation)Bil-GAT · TrafficRMSE: 0.062 -> 0.057Improved
add group-action basis modeling to attention encoder (ablation)Bil-GAT · PowerRMSE: 0.029 -> 0.026Improved
add implicit bilevel optimization (ablation)GAT · PowerRMSE: 0.065 -> 0.026Improved
add implicit bilevel optimization (ablation)GAT · TrafficRMSE: 0.067 -> 0.057Improved
increase basis size from 64 to 256 (ablation)FlowSymm · TrafficRMSE: 0.060 -> 0.057Improved
increase basis size from 256 to 512 (ablation)FlowSymm · Traffic, Power, BikeRMSE: 0.057 -> 0.057No clear effect
learn coefficients over valid basis instead of predicting then projecting (ablation)FlowSymm · TrafficRMSE: 0.201 -> 0.057Improved
replace Bil-GCN diagonal regularizer with FlowSymm group actionsBil-GCN · TrafficRMSE: 0.062 -> 0.057Improved
replace Bil-GCN diagonal regularizer with FlowSymm group actionsBil-GCN · TrafficMAE: 0.034 -> 0.028Improved
add 50% Gaussian noise to injection vector c at test timeFlowSymm · TrafficRMSE: 0.057 -> 0.060Hurt
add 50% Gaussian noise to injection vector c at test timeFlowSymm · PowerRMSE: 0.026 -> 0.028Hurt

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