Research map / Reinforcement learning / Visual and representation RL
RSPlace: Rotation Sensing Macro Placement via Bidirectional Tree Expansion
Tianyi Liu, Yaxin Xu, Lin Geng, Ningzhong Liu, Han Sun, Yu Wang
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 |
|---|---|---|---|
| Add rotation-aware sensing, perturbation and search (RSPlace) over EfficientPlace | EfficientPlace · ISPD2005 adaptec1 | HPWL (x10^5): 5.90 -> 4.72 | Improved |
| Use RSPlace instead of WireMask-EA | WireMask-EA · ISPD2005 bigblue4 | HPWL (x10^5): 86.88 -> 82.29 | Improved |
| Use RSPlace instead of EfficientPlace | EfficientPlace · ISPD2005 adaptec2 | HPWL (x10^5): 42.52 -> 39.79 | Improved |
| Use RSPlace instead of EfficientPlace | EfficientPlace · ISPD2005 adaptec3 | HPWL (x10^5): 55.75 -> 54.50 | Improved |
| Use RSPlace instead of EfficientPlace | EfficientPlace · ISPD2005 bigblue1 | HPWL (x10^5): 2.18 -> 2.03 | Improved |
| Use RSPlace instead of EfficientPlace | EfficientPlace · ISPD2005 bigblue2 | HPWL (x10^5): 4.19 -> 3.99 | Improved |
| Use RSPlace instead of EfficientPlace | EfficientPlace · ISPD2005 adaptec2 visualization | HPWL (x10^5): 43.35 -> 38.00 | Improved |
| Add guided rotation sensing instead of random rotation of macros (ablation) | RSPlace · ISPD2005 adaptec1 | HPWL | Improved |
| Add rotation of macros (ablation) | RSPlace · ISPD2005 adaptec1 | HPWL | Improved |
| Add guided top-down expansion instead of random perturbation only (ablation) | RSPlace · ISPD2005 adaptec1 | HPWL | Improved |
| Add bottom-up expansion with placement perturbation (ablation) | RSPlace · ISPD2005 adaptec1 | HPWL | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MaskPlace (3k) — MaskPlace: Fast Chip Placement via Reinforced Visual Representation Learning (NeurIPS 2022) · better on 8 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Reinforcement Learning within Tree Search for Fast Macro Placement (ICML 2024 · similarity 0.93)
- Reinforcement Learning Policy as Macro Regulator Rather than Macro Placer (NeurIPS 2024 · similarity 0.88)
- Chip Placement with Diffusion Models (ICML 2025 · similarity 0.88)
- See it to Place it: Evolving Macro Placements with Vision Language Models (NeurIPS 2026 · similarity 0.87)
- EGPlace: An Efficient Macro Placement Method via Evolutionary Search with Greedy Repositioning Guided Mutation (ICML 2025 · similarity 0.86)
- FlexPlanner: Flexible 3D Floorplanning via Deep Reinforcement Learning in Hybrid Action Space with Multi-Modality Representation (NeurIPS 2024 · 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.
