Research map / Vision-language and multimodal / Cross-modal learning
Topology Matters in RTL Circuit Representation Learning
Mingyu Zhao, Xun He, Jiawei Liu, Jianwang Zhai, Chuan Shi
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 TopoRTL topology-aware multimodal model instead of text-only LLM embeddings | TopoRTL · PPA prediction: Area | PCC: up 5.5% Area PCC | Improved |
| use TopoRTL instead of best baselines for power | TopoRTL · PPA prediction: Power | MAPE: down 31.5% Power MAPE | Improved |
| use TopoRTL instead of CircuitFusion | TopoRTL · PPA prediction: Area | PCC: 0.647 -> 0.863 | Improved |
| use TopoRTL instead of CircuitFusion with fewer parameters | TopoRTL · PPA model size | parameters: 150.59M -> 29.13M | Improved |
| use TopoRTL for timing slack versus best text model | TopoRTL · PPA prediction: Slack | PCC: 0.909 vs 0.909 | No clear effect |
| add TopoRTL for WNS prediction | TopoRTL · PPA prediction: WNS | RRSE: 0.580 | Improved |
| add topology-aware bit-width positional encoding (ablation) | TopoRTL · PPA and retrieval | PCC/AUC | Improved |
| add max-path and density encodings (ablation) | TopoRTL · PPA and retrieval | PCC/AUC | Mixed |
| add topology-guided cross-modal alignment loss (ablation) | TopoRTL · PPA and retrieval | PCC/AUC | Mixed |
| use TopoRTL for circuit retrieval versus all baselines | TopoRTL · Circuit retrieval (L=5 to 15) | AUC: near 0.8 AUC | Improved |
Closest papers by meaning
- CircuitFusion: Multimodal Circuit Representation Learning for Agile Chip Design (ICLR 2025 · similarity 0.88)
- Bridging Layout and RTL: Knowledge Distillation based Timing Prediction (ICML 2025 · similarity 0.88)
- AnalogToBi: Device-Level Analog Circuit Topology Generation via Bipartite Graph and Grammar Guided Decoding (NeurIPS 2026 · similarity 0.86)
- DeepLayout: Learning Neural Representations of Circuit Placement Layout (ICML 2025 · similarity 0.86)
- SymRTLO: Enhancing RTL Code Optimization with LLMs and Neuron-Inspired Symbolic Reasoning (NeurIPS 2025 · similarity 0.85)
- CircuitSense: A Hierarchical Circuit System Benchmark Bridging Visual Comprehension and Symbolic Reasoning in Engineering Design Process (ICLR 2026 · similarity 0.85)
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.
