# AttentionCap: Transformer Based Capacitance Matrix Learning Toward Full-Chip Extraction Source: [AttentionCap: Transformer Based Capacitance Matrix Learning Toward Full-Chip Extraction]() Collection index: [llms.txt]() Data updated: 2026-06-10T05:00:46.690Z Authors: Jiechen Huang, Hector R. Rodriguez, Dingcheng Yang, Zuochang Ye, Yibo Lin, Wenjian Yu Published: 2026-06-06T13:20:09.000Z ## Abstract As capacitance extraction accuracy of rule-based pattern matching becomes difficult to sustain at advanced nodes, a growing trend emerges to develop deep-learning-based 2D capacitance models. However, existing MLP- and CNN-based methods constrain their input to fixed metal-layer combinations in a specific process node, limiting their usability in practice. Recognizing the inherent similarity between capacitance matrix and the prevailing attention mechanism, we propose AttentionCap, a customized Transformer for capacitance matrix learning, with a Gram representation framework, a physics-aligned symmetric-attention output layer, and a novel normalized Laplacian loss. We also introduce a process-node embedding to enable multi-node learning. Trained on synthetic data, AttentionCap attains 0.67\\%/3.99\\% self/coupling-capacitance error on unseen real designs under a multi-layer and multi-node setting, surpassing the CNN-Cap baseline with 4.6$\\times$/5.7$\\times$ lower self/coupling error and 192$\\times$ faster inference speed. A pretrained AttentionCap accurately transfers to an unseen node with only 5K samples and 4K finetuning steps. With sufficient accuracy on unseen real designs and strong transferability to new process nodes, AttentionCap offers highly practical value for modern EDA workflows. Code and data are available at https://github.com/THU-numbda/AttentionCap. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Capacitance Extraction on CNN-Cap Pattern-C 55nm (val)](): Capacitance Extraction; CNN-Cap Pattern-C 55nm (val); Error Total: 3. - [Capacitance Extraction on CNN-Cap Pattern-C 15nm (val)](): Capacitance Extraction; CNN-Cap Pattern-C 15nm (val); Total Error (Errtot): 2. - [Capacitance Extraction on Synthetic data 65nm (val)](): Capacitance Extraction; Synthetic data 65nm (val); Total Error: 0.85. - [Capacitance Extraction on Real-design 65nm (test)](): Capacitance Extraction; Real-design 65nm (test); Total Error: 0.98. - [Capacitance Extraction on Synthetic data 7nm (val)](): Capacitance Extraction; Synthetic data 7nm (val); Total Error: 0.47. - [Capacitance Extraction on Real-design 7nm (test)](): Capacitance Extraction; Real-design 7nm (test); Total Error: 1.52. - [Capacitance Extraction on Synthetic data Mixed nodes (val)](): Capacitance Extraction; Synthetic data Mixed nodes (val); Total Error: 0.44. - [Capacitance Extraction on Real-design Mixed nodes (test)](): Capacitance Extraction; Real-design Mixed nodes (test); Total Error: 0.67. ## Original sources and implementation [Original paper]() [PDF]()