MacroAgent: Regularity-Aware Macro Legalization with LLM-Agent-Designed Contour Algorithms

• URL: http://arxiv.org/abs/2608.24946v1
• Date: Mon, 24 Aug 2026 15:31:52 GMT
• Title: MacroAgent: Regularity-Aware Macro Legalization with LLM-Agent-Designed Contour Algorithms
• Abstract summary: We present a novel framework for macro legalization using Large Language Models (LLMs)<n>Compared with state-of-the-art macro legalization works, experimental results on TILOS and Chipyard benchmarks demonstrate a 2 to 8 fold improvement in layout regularity.<n>End-to-end evaluation through Cadence Innovus place-and-route confirms that the regularity improvements translate into tangible PPA gains.
• Score: 21.59953270787966
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Macros constitute a large part of the core area in modern very large-scale integration (VLSI) designs. Moreover, macro positions have a significant impact on the final quality of result (QoR), and macro legalization is typically the final step in determining the macro positions. However, existing approaches related to macro legalization either lack robustness or incur substantial computational costs or neglect the regularity between macros. To address these limitations, we introduce MacroAgent. The novel framework is a four-stage approach: clustering, contour generation, template matching, and inter-cluster refinement. We propose leveraging Large Language Models (LLMs) to discover multiple, effective heuristic regularity-aware contour algorithms. This framework successfully generates robust and effective algorithmic solutions for macro legalization. Compared with state-of-the-art macro legalization works, experimental results on TILOS and Chipyard benchmarks demonstrate a 2 to 8 fold improvement in layout regularity, a 3% to 5% reduction in routed wirelength with comparable congestion after global routing, and significantly better robustness with an acceptable runtime. Furthermore, end-to-end evaluation through Cadence Innovus place-and-route confirms that the regularity improvements translate into tangible PPA gains, including 2.9% lower routed wirelength and 68.3% TNS improvement over the DREAMPlace macro legalization baseline; it also achieves 1.8% lower routed wirelength when integrated into the Innovus macro placement flow.

Related papers

• ROSETTA: Efficient and Accurate Privacy-Preserving LLM Decoding via Hybrid CKKS/TFHE Evaluation [9.045545175248671]
  Generative large language models (LLMs) have achieved state-of-the-art performance on many real-world tasks such as code generation and question answering.<n>Their pervasive deployment raises serious privacy concerns, motivating private inference frameworks based on fully homomorphic encryption (FHE)
  arXiv  Detail & Related papers  (2026-09-15T09:51:33Z)
• LEAP: Learnable End-to-End Adaptive Pruning of Large Language Models [19.274512633962086]
  State-of-the-art methods for unstructured pruning are layer-wise surrogates derived from the Optimal Brain Surgeon principle.<n>We introduce LEAP, which replaces this intractable parameterization with a per-weight Bernoulli-via-Gumbel sigmoid relaxation.<n>Across five LLM families from 0.5B to 8B parameters at 50% and 60% sparsity, LEAP improves six-task average zero-shot accuracy by +2.59 points on average over ADMM.
  arXiv  Detail & Related papers  (2026-05-17T07:01:13Z)
• Enhancing Multilingual Counterfactual Generation through Alignment-as-Preference Optimization [14.36297254909447]
  We introduce Macro, a preference alignment framework that applies Direct Preference Optimization to multilingual SCE generation.<n>We show that Macro improves validity by 12.55% on average over the chain-of-thought baseline without degrading minimality.<n>Our results highlight preference optimization as a promising direction for enhancing multilingual model explanations.
  arXiv  Detail & Related papers  (2026-05-12T06:56:18Z)
• MAny: Merge Anything for Multimodal Continual Instruction Tuning [52.50936513604062]
  textbfMAny (textbfMAny) is a framework that merges task-specific knowledge through textbfCross-modal textbfProjection textbfMerging.<n>textbfLow-rank textbfParameter textbfMerging (textbfLPM)
  arXiv  Detail & Related papers  (2026-04-15T15:57:23Z)
• PanFoMa: A Lightweight Foundation Model and Benchmark for Pan-Cancer [54.958921946378304]
  We introduce PanFoMa, a lightweight hybrid neural network that combines the strengths of Transformers and state-space models.<n>PanFoMa consists of a front-end local-context encoder with shared self-attention layers to capture complex, order-independent gene interactions.<n>We also construct a large-scale pan-cancer single-cell benchmark, PanFoMaBench, containing over 3.5 million high-quality cells.
  arXiv  Detail & Related papers  (2025-12-02T08:31:31Z)
• Re$^{\text{2}}$MaP: Macro Placement by Recursively Prototyping and Packing Tree-based Relocating [67.49674976434322]
  This work introduces the Re$text2$MaP method, which generates expert-quality macro placements.<n>We use DREAMPlace to build a mixed-size placement prototype and obtain reference positions for each macro and cluster.<n>A packing tree-based relocating procedure is then designed to jointly adjust the locations of macro groups and the macros within each group.
  arXiv  Detail & Related papers  (2025-11-11T09:56:10Z)
• Non-Overlapping Placement of Macro Cells based on Reinforcement Learning in Chip Design [20.500468654567033]
  This paper proposes an end-to-end placement method, SRLPlacer, based on reinforcement learning. By assessing on the public benchmark ISPD2005, the proposed SRLPlacer can effectively solve the overlap problem between macro cells.
  arXiv  Detail & Related papers  (2024-07-26T04:15:54Z)
• DB-LLM: Accurate Dual-Binarization for Efficient LLMs [83.70686728471547]
  Large language models (LLMs) have significantly advanced the field of natural language processing. Existing ultra-low-bit quantization always causes severe accuracy drops. We propose a novel Dual-Binarization method for LLMs, namely DB-LLM.
  arXiv  Detail & Related papers  (2024-02-19T09:04:30Z)
• Generative and Contrastive Paradigms Are Complementary for Graph Self-Supervised Learning [56.45977379288308]
  Masked autoencoder (MAE) learns to reconstruct masked graph edges or node features. Contrastive Learning (CL) maximizes the similarity between augmented views of the same graph. We propose graph contrastive masked autoencoder (GCMAE) framework to unify MAE and CL.
  arXiv  Detail & Related papers  (2023-10-24T05:06:06Z)
• Squeezeformer: An Efficient Transformer for Automatic Speech Recognition [99.349598600887]
  Conformer is the de facto backbone model for various downstream speech tasks based on its hybrid attention-convolution architecture. We propose the Squeezeformer model, which consistently outperforms the state-of-the-art ASR models under the same training schemes.
  arXiv  Detail & Related papers  (2022-06-02T06:06:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.