AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

FusionCell: Cross-Attentive Fusion of Layout Geometry and Netlist Topology for Standard-Cell Performance Prediction

Haoyi Zhang, Kairong Guo, Bojie Zhang, Yibo Lin, Runsheng Wang

ICML 2026 · Graph learning › Knowledge and temporal graphs · approach: Heterogeneous graph learning · 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 correlation edges between related nets (ablation)FusionCell · ASAP7 standard-cell datasetaverage MAPE: 1.45% -> 0.92%Improved
add correlation edges between related nets (ablation)FusionCell · ASAP7 standard-cell datasetaverage Spearman rho: 0.82 -> 0.86Improved
add correlation edges between related nets (ablation)FusionCell · ASAP7 standard-cell datasetaverage Kendall tau: 0.73 -> 0.76Improved
add correlation edges between related nets (ablation)FusionCell · ASAP7 standard-cell datasetaverage R2: 0.975 -> 0.977Improved
replace symmetrical bidirectional cross-attention with graph-query layout-key cross-attention (ablation)dual-stream fusion · ASAP7 standard-cell datasetaverage MAPE: 2.19% -> 0.92%Improved
replace symmetrical bidirectional cross-attention with graph-query layout-key cross-attention (ablation)dual-stream fusion · ASAP7 standard-cell datasetaverage Spearman rho: 0.81 -> 0.86Improved
replace late fusion (concatenated pooled vectors) with graph-query layout-key cross-attention (ablation)dual-stream model · ASAP7 standard-cell datasetaverage MAPE: 2.33% -> 0.92%Improved
use symmetrical fusion instead of late fusion (ablation)dual-stream model · ASAP7 standard-cell datasetaverage MAPE: 2.33% -> 2.19%No clear effect
add netlist graph input to layout-only model with topology-guided fusion (ablation)DeiT vision-only · ASAP7 standard-cell datasetaverage MAPE: 3.16% -> 0.92%Improved
add netlist graph input to layout-only model with topology-guided fusion (ablation)DeiT vision-only · ASAP7 standard-cell datasetaverage Spearman rho: 0.56 -> 0.86Improved
replace coarse layout-augmented graph (ProtoCellLayout) with explicit DeiT layout encoding and cross-attentionlayout representation · ASAP7 standard-cell datasetaverage MAPE: 4.16% -> 0.92%Improved
replace coarse layout-augmented graph (ProtoCellLayout) with explicit DeiT layout encoding and cross-attentionlayout representation · ASAP7 standard-cell datasetaverage Spearman rho: 0.30 -> 0.86Improved

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