SynCircuit: Automated Generation of New Synthetic RTL Circuits Can Enable Big Data in Circuits

2025-08-26

AI Summary

AI-assisted IC design is hindered by the severe scarcity of publicly available circuit data. To address this, we propose the first diffusion-based generative model tailored for directed cyclic graphs (DCGs), integrating circuit-constrained refinement and Monte Carlo Tree Search (MCTS)-guided optimization to eliminate logical redundancy—enabling fully automated, functionally correct, and structurally realistic RTL-level circuit synthesis. Generated circuits are rigorously validated via HDL modeling and formal logic equivalence checking to ensure correctness and fidelity, enabling construction of a large-scale, high-quality synthetic circuit dataset. Experiments demonstrate that the synthesized circuits exhibit distributional alignment with real-world designs and significantly improve generalization and accuracy of downstream ML tasks—including logic synthesis prediction and power estimation. This work pioneers the application of diffusion models to circuit graph generation and establishes a scalable, data-driven foundation for next-generation IC design automation.

Technology Category

• Machine Learning: Probabilistic Circuits and Graphical Models
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

In recent years, AI-assisted IC design methods have demonstrated great potential, but the availability of circuit design data is extremely limited, especially in the public domain. The lack of circuit data has become the primary bottleneck in developing AI-assisted IC design methods. In this work, we make the first attempt, SynCircuit, to generate new synthetic circuits with valid functionalities in the HDL format. SynCircuit automatically generates synthetic data using a framework with three innovative steps: 1) We propose a customized diffusion-based generative model to resolve the Directed Cyclic Graph (DCG) generation task, which has not been well explored in the AI community. 2) To ensure our circuit is valid, we enforce the circuit constraints by refining the initial graph generation outputs. 3) The Monte Carlo tree search (MCTS) method further optimizes the logic redundancy in the generated graph. Experimental results demonstrate that our proposed SynCircuit can generate more realistic synthetic circuits and enhance ML model performance in downstream circuit design tasks.

Problem

• Automatically generates synthetic RTL circuits with valid HDL functionalities
• Addresses limited circuit design data bottleneck for AI-assisted IC development
• Resolves directed cyclic graph generation using customized diffusion models

Innovation

• Diffusion model generates directed cyclic graphs
• Constraint refinement ensures circuit validity
• Monte Carlo tree search optimizes logic redundancy