Research map / Point clouds / Point cloud classification
Circuit as Set of Points
Jialv Zou, Xinggang Wang, JiaHao Guo, Wenyu Liu, Qian Zhang, Chang Huang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use CircuitFormer point-cloud features instead of hand-crafted CNN input features | Congestion prediction · CircuitNet | Pearson correlation: 0.6085 -> 0.6374 | Improved |
| Use CircuitFormer point-cloud features instead of hand-crafted CNN input features | Congestion prediction · ISPD2015 | Spearman correlation: 0.1575 -> 0.1797 | Improved |
| Use raw point-cloud features with CircuitFormer rather than RouteNet's hand-crafted features | DRC violation prediction · CircuitNet | Pearson correlation: 0.4698 -> 0.3707 | Hurt |
| Use raw point-cloud features with CircuitFormer rather than RouteNet's hand-crafted features | DRC violation prediction · CircuitNet | Spearman correlation: 0.2782 -> 0.3094 | Improved |
| Replace the proposed encoder with point-based perception encoders (ablation) | CircuitFormer · CircuitNet congestion | Pearson correlation: 0.6374 -> 0.4086 | Hurt |
| Replace the proposed encoder with PointPillar (ablation) | CircuitFormer · CircuitNet congestion | Pearson correlation: 0.6374 -> 0.6151 | Hurt |
| Add cross-grid information exchange through the proposed attention encoder rather than PointPillar's within-grid features (ablation) | CircuitFormer · CircuitNet congestion | Pearson correlation: 0.6151 -> 0.6374 | Improved |
| Add label distribution smoothing (ablation) | CircuitFormer · CircuitNet congestion | Spearman correlation: 0.5143 -> 0.5282 | Improved |
| Add label distribution smoothing (ablation) | CircuitFormer · CircuitNet DRC violation | Pearson correlation: 0.3552 -> 0.3707 | Improved |
| Increase encoder depth from one GA stage to four (ablation) | CircuitFormer · CircuitNet congestion | Pearson correlation: 0.5260 -> 0.6374 | Improved |
| Add the ResNet-18 backbone before the UNet++ head (ablation) | CircuitFormer decoder · CircuitNet congestion | Pearson correlation: 0.3574 -> 0.6374 | Improved |
| Replace UNet++ with UNet as the segmentation head (ablation) | CircuitFormer decoder · CircuitNet congestion | Pearson correlation: 0.6374 -> 0.5237 | Hurt |
Closest papers by meaning
- DeepLayout: Learning Neural Representations of Circuit Placement Layout (ICML 2025 · similarity 0.85)
- DeepGate4: Efficient and Effective Representation Learning for Circuit Design at Scale (ICLR 2025 · similarity 0.85)
- CircuitFusion: Multimodal Circuit Representation Learning for Agile Chip Design (ICLR 2025 · similarity 0.84)
- Circuit Representation Learning with Masked Gate Modeling and Verilog-AIG Alignment (ICLR 2025 · similarity 0.84)
- CircuitSense: A Hierarchical Circuit System Benchmark Bridging Visual Comprehension and Symbolic Reasoning in Engineering Design Process (ICLR 2026 · similarity 0.83)
- Benchmarking End-To-End Performance of AI-Based Chip Placement Algorithms (NeurIPS 2025 · similarity 0.83)
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.
