AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Bures-Wasserstein Flow Matching for Graph Generation

Keyue Jiang, Jiahao Cui, Xiaowen Dong, Laura Toni

ICLR 2026 · Graph learning › Graph self-supervised learning · approach: Diffusion-based graph generation

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 linear interpolation with Bures-Wasserstein interpolation in flow matchingBWFlow · Planar graphsV.U.N. and A.Ratio: V.U.N. 77.5 -> 84.8, A.Ratio 3.5 -> 2.4 vs DeFoGImproved
use BW interpolation on tree graphsBWFlow · Tree graphsV.U.N.: 81.5 vs 84.5 (Cometh)No clear effect
use BW interpolation on tree graphsBWFlow · Tree graphsA.Ratio: 1.9 -> 1.3 vs DeFoGImproved
use BW interpolation, with 30 instead of 1k sampling stepsBWFlow · Planar (30 sampling steps)V.U.N.: 72.0 -> 77.0 vs DeFoG-2; 24.5 for DeFoG-1Improved
use BW interpolation, with 30 sampling stepsBWFlow · SBM (30 sampling steps)V.U.N.: 47.5 -> 52.0 vs DeFoG-2Improved
add path manipulation (time distortion, target guidance) to linear-interpolation flow (ablation)DeFoG · Planar (30 sampling steps)V.U.N.: 24.5 -> 72.0Improved
use BW interpolation instead of linear, harmonic or geometric interpolation (ablation)BWFlow · Plain graphsV.U.N.Improved
use BW interpolation instead of linear interpolation (ablation)BWFlow · Planar training convergenceconvergence speedImproved
use harmonic or geometric interpolation (ablation)BWFlow · SBMpath validityHurt
use BWFlow for 3D molecule generation without bond typesBWFlow · QM9 with hMol.Stab.: 93.98 -> 97.84 vs MiDiImproved
use BWFlow for 3D molecule generation without bond typesBWFlow · GEOM-DRUGS with hConnected (%): 79.15 -> 73.53 vs MiDiHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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