LaMAGIC2: Advanced Circuit Formulations for Language Model-Based Analog Topology Generation

2025-06-11

AI Summary

Existing approaches to automated analog circuit topology synthesis suffer from high token complexity (O(|V|²)), poor numerical precision sensitivity, and weak cross-scale generalization. To address these limitations, this paper proposes an efficient generative framework leveraging large language models (LLMs). Its core innovation is the Sparse Floating-point Circuit Input (SFCI) specification—a novel, compact input representation that integrates identifier-enhanced device recognition with floating-point numerical embedding, reducing token complexity to O(|V|) while significantly improving numerical modeling accuracy and structural generalization. By incorporating circuit graph structural normalization and supervised fine-tuning, the method achieves a 34% increase in success rate under a stringent 0.01 tolerance and reduces mean squared error by one order of magnitude. Furthermore, it demonstrates superior transferability to large-scale circuits, with migration performance improvements up to 58.5%.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Probabilistic Circuits and Graphical Models
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Automation of analog topology design is crucial due to customized requirements of modern applications with heavily manual engineering efforts. The state-of-the-art work applies a sequence-to-sequence approach and supervised finetuning on language models to generate topologies given user specifications. However, its circuit formulation is inefficient due to O(|V |2) token length and suffers from low precision sensitivity to numeric inputs. In this work, we introduce LaMAGIC2, a succinct float-input canonical formulation with identifier (SFCI) for language model-based analog topology generation. SFCI addresses these challenges by improving component-type recognition through identifier-based representations, reducing token length complexity to O(|V |), and enhancing numeric precision sensitivity for better performance under tight tolerances. Our experiments demonstrate that LaMAGIC2 achieves 34% higher success rates under a tight tolerance of 0.01 and 10X lower MSEs compared to a prior method. LaMAGIC2 also exhibits better transferability for circuits with more vertices with up to 58.5% improvement. These advancements establish LaMAGIC2 as a robust framework for analog topology generation.

Problem

• Inefficient O(|V|²) token length in analog topology generation
• Low precision sensitivity to numeric inputs in circuits
• Poor component-type recognition in existing formulations

Innovation

• SFCI formulation reduces token length complexity
• Identifier-based representations improve component recognition
• Enhanced numeric precision sensitivity for tight tolerances