Autonomous Evolution of EDA Tools: Multi-Agent Self-Evolved ABC

2026-04-16

AI Summary

This work proposes the first self-evolving logic synthesis framework that overcomes the limitations of traditional electronic design automation (EDA) tools, which rely on hand-crafted heuristics and lack the ability to autonomously and continuously improve synthesis performance. Operating within the fixed single-binary architecture of ABC, the framework leverages a multi-agent system powered by large language models to iteratively rewrite code submodules through programming-guided prompts. End-to-end autonomous evolution across the entire codebase is achieved by integrating functional correctness verification with quality-of-results (QoR) feedback. Demonstrating unprecedented capability, the approach enables fully automated strategy discovery and performance enhancement in an industrial-scale EDA tool comprising millions of lines of code, consistently improving QoR across standard benchmarks including ISCAS, VTR, EPFL, and IWLS—thereby validating its capacity for incremental, self-directed optimization of synthesis quality.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Search and Optimization: Evolutionary Computation
• Machine Learning: Evolutionary Learning

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

This paper introduces the first \emph{self-evolving} logic synthesis framework, which leverages Large Language Model (LLM) agents to autonomously improve the source code of \textsc{ABC}, the widely adopted logic synthesis system. Our framework operates on the \emph{entire integrated ABC codebase}, and the output repository preserves its single-binary execution model and command interface. In the initial evolution cycle, we bootstrap the system using existing prior open-source synthesis components, covering flow tuning, logic minimization, and technology mapping, but without manually injecting new heuristics. On top of this foundation, a team of LLM-based agents iteratively rewrites and evolves specific sub-components of ABC following our ``programming guidance`` prompts under a unified correctness and QoR-driven evaluation loop. Each evolution cycle proposes code modifications, compiles the integrated binary, validates correctness, and evaluates quality-of-results (QoR) on \emph{multi-suite benchmarks including ISCAS~85/89/99, VTR, EPFL, and IWLS~2005}. Through continuous feedback, the system discovers optimizations beyond human-designed heuristics, effectively \emph{learning new synthesis strategies} that enhance QoR. We detail the architecture of this self-improving system, its integration with \textsc{ABC}, and results demonstrating that the framework can autonomously and progressively improve EDA tool at full million-line scale.

Problem

• Autonomous Evolution
• Logic Synthesis
• EDA Tools
• Self-Evolving Systems
• Quality-of-Results (QoR)

Innovation

• self-evolving
• LLM agents
• logic synthesis
• quality-of-results (QoR)
• multi-agent system