---
title: "Physics-Guided Geometric Diffusion for Macro Placement Generation"
authors:
  - "Jongho Yoon"
  - "Jinsung Jeon"
  - "Seokhyeong Kang"
doi: "10.24963/ijcai.2026/735"
url: "https://ijcai.org/proceedings/2026/735"
pdf: "https://ijcai.org/proceedings/2026/0735.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "6609-6616"
track: "AI4Tech: AI Enabling Technologies"
keywords:
  - "Domain-specific AI4Tech: AI4Manufacturing"
  - "Emerging AI4Tech : Emerging AI4Tech areas"
  - "AI4Tech infrastructure/systems: AI chips, AI sensors, AI computers"
  - "Advanced AI4Tech: Generative and LLMs-driven AI4Tech"
---

# Physics-Guided Geometric Diffusion for Macro Placement Generation

Jongho Yoon, Jinsung Jeon, Seokhyeong Kang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 6609-6616. DOI: [10.24963/ijcai.2026/735](https://doi.org/10.24963/ijcai.2026/735)

## Abstract

Macro placement is a pivotal stage in VLSI physical design, fundamentally determining the overall chip performance. Recent data-driven placement methods have demonstrated significant potential, yet they often struggle to handle sequential dependencies and to balance topological connectivity with physical constraints. To bridge this gap, we propose MacroDiff+, a physics-guided geometric diffusion framework. Specifically, we design a dual-domain denoising architecture that couples topological connectivity encoded by heterogeneous GNNs with global geometric context modeled by a Transformer. Furthermore, we introduce Physics-Guided Sampling, an inference strategy that actively steers the generation using explicit gradients to ensure both statistical plausibility and physical validity. On the ISPD2005 MMS benchmarks, MacroDiff+ outperforms state-of-the-art baselines with a 6.1–6.2% reduction in wirelength. Notably, it exhibits superior stability and scalability on large-scale designs where prior methods fail to converge. The source code is provided at https://github.com/jhy00n/MacroDiff-plus.