AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Beyond Tokens: Enhancing RTL Quality Estimation via Structural Graph Learning

Yi Liu, Hongji Zhang, Yiwen Wang, Dimitris Tsaras, Lei Chen, Mingxuan Yuan, Qiang Xu

ICML 2026 · Graph learning › Graph self-supervised learning · approach: Brain and biological graph learning

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
remove structure-aware masked node modeling (ablation)StructRTL · area prediction, w/o KDR2: 0.7249 -> 0.7463Improved
add edge prediction pretraining (ablation)StructRTL · area prediction, w/o KDR2: 0.7018 -> 0.7463Improved
add knowledge distillation from post-mapping netlist predictorStructRTL · delay predictionR2: 0.7630 -> 0.8872Improved
add knowledge distillation from post-mapping netlist predictorStructRTL · area predictionR2: 0.7463 -> 0.8676Improved
use CDFG graph self-supervised model instead of Verilog-LLM embeddingsCodeV-QW-7B · area prediction, w/o KDR2: 0.6353 -> 0.7463Improved
mask post-GNN embeddings instead of raw input features (ablation)StructRTL · area and delay predictiondownstream R2Improved
add global positional embeddings (ablation)StructRTL · area and delay predictiondownstream R2Improved
combine GNN with Transformer instead of either alone (ablation)StructRTL · area and delay predictiondownstream R2Improved
add stratified masking (ablation)StructRTL · masked node modelingoverall validation accuracy: 82.44% -> 82.27% overall; rare operators improveMixed
vary edge prediction sampling ratio from 10% to 40% (ablation)StructRTL · area and delay R2R2: 0.7448 to 0.7463 (area)No clear effect
feed textualized CDFG (.dot) to LLM instead of RTL codeVeriDistill CodeV models · area prediction, w/o KDR2: 0.6353 -> 0.5639 (CodeV-QW-7B)Hurt
fully fine-tune LLM on RTL code for quality estimationCodeV-QW-7B · area validation, w/o KDR2: 0.6353 -> 0.2685Hurt

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