EvoDRC: A Self-Evolving Agentic Framework for Automated DRC Violation Repair

2026-07-22

AI Summary

This work addresses the inefficiency and heavy reliance on manual intervention in fixing design rule check (DRC) violations at advanced process nodes by proposing a self-evolving repair framework powered by large language model (LLM) agents. The approach partitions layouts into bounded repair regions and integrates local DRC analysis, connectivity verification, and impact preview tools to enable automated correction. It introduces, for the first time, skill distillation and continuous evolution mechanisms to facilitate cross-design knowledge transfer and online accumulation of repair expertise. A modular agent architecture ensures both the effectiveness and safety of the repairs. Evaluated on seven modules from the DAC26 DRC benchmark, the method achieves an overall violation reduction rate of 73.5%, significantly outperforming existing baseline approaches.

Abstract

Design rule check (DRC) closure remains a major bottleneck in advanced-node physical design. Although detailed routers are rule-aware, residual design rule violations (DRVs) often require manual engineering change order iterations. Automating this process is challenging because repairs must account for complex geometric interactions, preserve circuit connectivity, and avoid introducing new violations. We present EvoDRC, a skill-evolution framework for agentic block-level DRC repair. EvoDRC initializes layer-specific repair skills using knowledge distilled from an unrelated reference design and continuously evolves these skills using traceable repair experience collected from the target design. EvoDRC decomposes the layout into bounded repair regions and assigns an LLM repair agent to each region. Local DRC analysis, connectivity-checking, and impact-preview tools provide feedback on proposed modifications. Repair operations and their resulting DRV changes are stored in a knowledge database and used to evolve the repair skills. Experiments on seven block-level designs from the DAC26 DRC Benchmark show that EvoDRC achieves a 73.5\% overall reduction compared to the reported baseline.

Problem

• Design Rule Check (DRC)
• DRC Violation Repair
• Physical Design
• Automated ECO
• Advanced-node

Innovation

• self-evolving
• agentic framework
• DRC violation repair
• skill distillation
• LLM-based repair agent