AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Generating Directed Graphs with Dual Attention and Asymmetric Encoding

Alba Carballo-Castro, Manuel Madeira, Yiming QIN, Dorina Thanou, Pascal Frossard

ICLR 2026 · Graph learning › Graph self-supervised learning · approach: Diffusion-based graph generation · 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
add dual attention instead of doubling network depth (RRWP) (ablation)Directo · ER-DAGV.U.N. (%): 72.0 -> 94.0Improved
add dual attention instead of doubling network depth (MagLap) (ablation)Directo · ER-DAGV.U.N. (%): 80.0 -> 91.0Improved
add dual attention instead of doubling network depth (RRWP) (ablation)Directo · SBMV.U.N. (%): 0.0 -> 99.5Improved
add dual attention instead of doubling network depth (MagLap) (ablation)Directo · SBMV.U.N. (%): 8.0 -> 77.0Improved
add dual attention instead of doubling network depth (RRWP) (ablation)Directo · SBMgraph statistics ratio: 26.3 -> 1.8Improved
add dual attention instead of doubling network depth (RRWP) (ablation)Directo · ER-DAGgraph statistics ratio: 4.9 -> 1.7Improved
add positional encodings (ablation)Directo · ER-DAG and SBMV.U.N. and RatioImproved
use direction-aware positional encodings instead of direction-agnostic Laplacian (ablation)Directo · ER-DAG and SBMV.U.N. and RatioImproved
use dual attention with no positional encoding (ablation)Directo · ER-DAG and SBMV.U.N.Improved
use Directo flow matching with RRWP instead of DeFoGDirecto RRWP · Visual GenomeV.U.N. (%): 50.8 -> 83.8Improved
use Directo MagLap instead of LayerDAGDirecto MagLap · TPU Tilesgraph statistics ratio: 413.6 -> 44.0Improved
increase training data amount (ablation)Directo · syntheticgeneration performanceMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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