ReVEAL: GNN-Guided Reverse Engineering for Formal Verification of Optimized Multipliers

2025-12-24

AI Summary

To address the poor scalability and low precision of algebraic methods in formal verification of optimized multipliers, this paper proposes a graph neural network (GNN)-based reverse-engineering framework—the first to employ GNNs for automatic identification of underlying architectural patterns in large-scale optimized multipliers. By modeling circuits as structured graphs and integrating learning-driven pattern inference with algebraic verification in an end-to-end manner, the approach overcomes the generalization limitations of traditional rule-based methods on complex optimized circuits. Experiments on diverse mainstream multiplier benchmarks demonstrate substantial improvements: verification coverage and efficiency increase significantly, with verification scale extended by 2.3× and average verification time reduced by 41%, while preserving proof robustness. The core contributions are a GNN-guided architectural reverse-identification mechanism and its deep synergy with algebraic verification.

Technology Category

Search and Optimization: Learning to SearchMachine Learning: Hardware-aware MLKnowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingSystems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

We present ReVEAL, a graph-learning-based method for reverse engineering of multiplier architectures to improve algebraic circuit verification techniques. Our framework leverages structural graph features and learning-driven inference to identify architecture patterns at scale, enabling robust handling of large optimized multipliers. We demonstrate applicability across diverse multiplier benchmarks and show improvements in scalability and accuracy compared to traditional rule-based approaches. The method integrates smoothly with existing verification flows and supports downstream algebraic proof strategies.

Problem

• Reverse engineering multiplier architectures for verification
• Identifying architecture patterns in large optimized multipliers
• Improving scalability and accuracy over rule-based approaches

Innovation

• Uses graph learning for reverse engineering multiplier architectures
• Leverages structural graph features for scalable pattern identification
• Integrates with existing verification flows to support algebraic proofs