Using GUI Agent for Electronic Design Automation

2025-12-12

AI Summary

Current GUI agents perform well in general-purpose software but exhibit weak capabilities in professional EDA/CAD tools, failing to replace human engineers. This work pioneers the systematic application of GUI agents to the EDA domain. We propose: (1) GUI-EDA—the first EDA-specific GUI benchmark dataset; (2) EDAgent—a dedicated evaluation framework for circuit design, integrating multi-tool screen understanding, action planning, and task reflection mechanisms; and (3) an end-to-end screenshot-to-action modeling paradigm tailored to the complexity of industrial-grade CAD interfaces. Evaluated across five EDA tools and five major physical design scenarios, our framework comprehensively benchmarks over 30 state-of-the-art GUI agents. EDAgent is the first GUI agent to outperform Ph.D. candidates in electrical engineering on real-world CAD tasks, achieving industrial usability.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Adversarial Agents
• Humans and AI: Intelligent User Interfaces

Application Category

• Economics, Online Markets and Human Computation: Economic aspects and information design of Web, GenAI, and cloud computing
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Graphical User Interface (GUI) agents adopt an end-to-end paradigm that maps a screenshot to an action sequence, thereby automating repetitive tasks in virtual environments. However, existing GUI agents are evaluated almost exclusively on commodity software such as Microsoft Word and Excel. Professional Computer-Aided Design (CAD) suites promise an order-of-magnitude higher economic return, yet remain the weakest performance domain for existing agents and are still far from replacing expert Electronic-Design-Automation (EDA) engineers. We therefore present the first systematic study that deploys GUI agents for EDA workflows. Our contributions are: (1) a large-scale dataset named GUI-EDA, including 5 CAD tools and 5 physical domains, comprising 2,000+ high-quality screenshot-answer-action pairs recorded by EDA scientists and engineers during real-world component design; (2) a comprehensive benchmark that evaluates 30+ mainstream GUI agents, demonstrating that EDA tasks constitute a major, unsolved challenge; and (3) an EDA-specialized metric named EDAgent, equipped with a reflection mechanism that achieves reliable performance on industrial CAD software and, for the first time, outperforms Ph.D. students majored in Electrical Engineering. This work extends GUI agents from generic office automation to specialized, high-value engineering domains and offers a new avenue for advancing EDA productivity. The dataset will be released at: https://github.com/aiben-ch/GUI-EDA.

Problem

• Automating EDA workflows with GUI agents
• Evaluating GUI agents on professional CAD software
• Creating specialized metrics for EDA automation performance

Innovation

• Deploying GUI agents for Electronic Design Automation workflows
• Creating a large-scale dataset with real-world screenshot-action pairs
• Developing a specialized metric with reflection mechanism for CAD software