CoEvoP&R: Co-Evolving Placement Objectives with Routing Feedback via Large Language Models

2026-07-19

AI Summary

This work addresses the optimization gap between traditional placement objectives—such as wirelength and density—and post-routing timing and congestion performance. It proposes the first integration of large language models (LLMs) into physical design by leveraging prompt engineering and evolutionary search within DREAMPlace to automatically discover differentiable, interpretable, and analytically embeddable placement objective functions. The method incorporates routing and timing proxies in a closed-loop feedback optimization framework. Evaluated on the ChiP-Bench and ICCAD 2015 benchmarks, it achieves an average post-routing wirelength reduction of 16.9%, congestion reduction of 36.7%, worst negative slack improvement of 0.70 ns, and total negative slack reduction of 912 ns, significantly narrowing the optimization disparity between placement and routing stages.

Abstract

Analytical placers rely on differentiable objective functions to guide placement, typically combining intermediate surrogate metrics such as half-perimeter wirelength (HPWL) and cell-density penalties. However, these placement-stage surrogates remain misaligned with downstream routed and timing quality. Prior work reduces this gap with human-designed terms or learned black-box surrogates, but the former requires expert retuning and the latter is difficult to explain, debug, or deploy in analytical placement flows. CoEvoP&R addresses these limitations with a large language model (LLM)-based framework that automatically evolves analytical placement objectives. At each generation, the prompt combines the restricted objective interface, baseline context, and archived prior candidates with routing-related feedback from placement, timing proxy, and routing tools. The LLM proposes readable differentiable objectives, which are embedded and validated in DREAMPlace, evaluated through a timing proxy and an actual router, and stored with their feedback to guide later generations. Across eight ChiP-Bench Nangate45 designs and three seeds, CoEvoP&R reduces post-route routed wirelength and congestion by 16.9% and 36.7%, with gains of 0.70 ns in worst negative slack and a 912 ns reduction in total negative slack magnitude over native DREAMPlace. Across eight ICCAD 2015 Superblue designs, it reduces post-route routed wirelength and congestion by 5.4% and 23.2%. Code is available at https://github.com/FCHXWH823/CoEvoP-R.git.

Problem

• placement
• routing
• objective function
• timing quality
• congestion

Innovation

• Large Language Models
• Analytical Placement
• Routing Feedback
• Co-Evolution
• Differentiable Objectives