Research map / Learning theory and algorithms / Combinatorial and search optimization
Macro Placement by Wire-Mask-Guided Black-Box Optimization
Yunqi Shi, Ke Xue, Lei Song, Chao Qian
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 WireMask-EA instead of MaskPlace | Macro placement · ISPD2005 adaptec1 | HPWL: 6.38 \pm 0.35 -> 5.91\pm0.07 | Improved |
| Use WireMask-EA instead of MaskPlace | Macro placement · ISPD2005 adaptec2 | HPWL: 73.75 \pm 6.35 -> 52.63\pm2.23 | Improved |
| Use WireMask-EA instead of MaskPlace | Macro placement · ISPD2005 adaptec3 | HPWL: 84.44 \pm 3.60 -> 57.75\pm1.16 | Improved |
| Use WireMask-EA instead of DREAMPlace | Macro placement · ISPD2005 adaptec4 | HPWL: 37.41\pm0.87 -> 58.79 \pm 1.02 | Hurt |
| Use WireMask-EA instead of MaskPlace | Macro placement · ISPD2005 bigblue1 | HPWL: 2.39 \pm 0.05 -> 2.12\pm0.01 | Improved |
| Use WireMask-EA instead of MaskPlace | Macro placement · ISPD2005 bigblue3 | HPWL: 91.11 \pm 7.83 -> 59.87\pm3.40 | Improved |
| Use WireMask-EA instead of MaskPlace | Macro placement · ISPD2005 bigblue4 | HPWL: 11.07 \pm 0.90 -> 8.28 \pm 0.25 | Improved |
| Use WireMask-EA instead of MaskPlace | Macro placement · ariane | HPWL: 14.63\pm0.20 -> 9.68\pm0.61 | Improved |
| Increase the grid partition number by 100 (ablation) | WireMask-EA · adaptec1 | HPWL: 5.91\pm0.07 -> 5.88\pm0.09 | No clear effect |
| Decrease the grid partition number by 100 (ablation) | WireMask-EA · adaptec2, adaptec3, adaptec4 | non-overlapping placement | Hurt |
| Use random macro-adjustment order instead of connected-cell-area order (ablation) | WireMask-EA · ISPD2005 macro-placement benchmarks | HPWL: 12.65\pm0.68 (adaptec1); \times (adaptec2, adaptec3, bigblue3, bigblue4) | Hurt |
| Order macros by their own area only instead of connected-cell area (ablation) | WireMask-EA · ISPD2005 macro-placement benchmarks | HPWL: 47.39\pm1.02 (adaptec2); 102.99\pm7.69 (bigblue3) | Mixed |
Closest papers by meaning
- See it to Place it: Evolving Macro Placements with Vision Language Models (NeurIPS 2026 · similarity 0.88)
- EGPlace: An Efficient Macro Placement Method via Evolutionary Search with Greedy Repositioning Guided Mutation (ICML 2025 · similarity 0.88)
- Chip Placement with Diffusion Models (ICML 2025 · similarity 0.86)
- Reinforcement Learning within Tree Search for Fast Macro Placement (ICML 2024 · similarity 0.85)
- Benchmarking End-To-End Performance of AI-Based Chip Placement Algorithms (NeurIPS 2025 · similarity 0.85)
- Reinforcement Learning Policy as Macro Regulator Rather than Macro Placer (NeurIPS 2024 · similarity 0.84)
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.
