AI Paper Atlas

Research map / Graph learning / GNN architectures

Pb4U-GNet: Resolution-Adaptive Garment Simulation via Propagation-before-Update Graph Network

Aoran Liu, Kun Hu, Clinton Ansun Mo, Qiuxia Wu, Wenxiong Kang, Zhiyong Wang

AAAI 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
add resolution-aware propagation control (adaptive message-passing steps) (ablation)Pb4U-GNet · VTO Lv.4 (38K), trained on 11K onlytotal physics loss: 1.08E+09 -> 2.22E-01Improved
add resolution-aware update scaling (ablation)Pb4U-GNet · VTO Lv.4 (38K), trained on 11K onlytotal physics loss: 7.34E+13 -> 2.22E-01Improved
add propagation control at training resolution (ablation)Pb4U-GNet · VTO Lv.1 (11K)total physics loss: -1.61E-03 (w/o) vs -1.66E-02No clear effect
use update scaling at training resolution (ablation)Pb4U-GNet · VTO Lv.1 (11K)total physics loss: -5.78E-03 (w/o) vs -1.66E-02No clear effect
use Pb4U-GNet vs baselines at unseen higher resolutionsPb4U-GNet · VTO Lv.4 (38K)total physics loss: 2.82E-01 (CCRAFT) -> 2.22E-01Improved
use Pb4U-GNet at training resolutionPb4U-GNet · VTO Lv.1 (11K)total physics loss: -1.66E-02 vs ESLR -2.56E-02No clear effect
adaptive propagation steps on low-resolution meshPb4U-GNet · VTO 12K dresslatency (ms): 50.0 ms vs MGN 46.4 ms, HOOD 50.8 msNo clear effect
generalize to unseen garment types (dress and cardigan)Pb4U-GNet · VTO unseen garments, highest resolutiontotal physics loss: 2.64E-01 (CCRAFT) -> 1.55E-01Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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