GRAIL: Graph Edit Distance and Node Alignment Using LLM-Generated Code

TL;DR

GRAIL leverages large language models and automated prompt tuning to generate interpretable programs for graph edit distance computation, overcoming data and generalization limitations of neural methods.

Contribution

It introduces a novel approach that uses LLM-generated code for GED calculation, eliminating the need for ground-truth data and enhancing interpretability and cross-domain adaptability.

Findings

Outperforms state-of-the-art GED approximation methods.

Demonstrates robust cross-domain generalization.

Provides interpretable, self-evolving GED computation programs.

Abstract

Graph Edit Distance (GED) is a widely used metric for measuring similarity between two graphs. Computing the optimal GED is NP-hard, leading to the development of various neural and non-neural heuristics. While neural methods have achieved improved approximation quality compared to non-neural approaches, they face significant challenges: (1) They require large amounts of ground truth data, which is itself NP-hard to compute. (2) They operate as black boxes, offering limited interpretability. (3) They lack cross-domain generalization, necessitating expensive retraining for each new dataset. We address these limitations with GRAIL, introducing a paradigm shift in this domain. Instead of training a neural model to predict GED, GRAIL employs a novel combination of large language models (LLMs) and automated prompt tuning to generate a program that is used to compute GED. This shift from predicting GED to generating programs imparts various advantages, including end-to-end interpretability and an autonomous self-evolutionary learning mechanism without ground-truth supervision. Extensive experiments on seven datasets confirm that GRAIL not only surpasses state-of-the-art GED approximation methods in prediction quality but also achieves robust cross-domain generalization across diverse graph distributions.