Adaptive Graph Refinement and Label Propagation with LLMs for Cost-Effective Entity Resolution

2026-05-25

AI Summary

This work addresses the limitations of traditional pipeline-based entity resolution—namely, graph sparsity, missing edges, and error propagation—by introducing Alper, a novel framework that unifies matching and clustering into a single co-optimization task. Alper iteratively performs probabilistic label propagation over a global dynamic graph, effectively integrating weak signals from graph neural propagation with strong signals from large language models (LLMs). It further incorporates an adaptive graph refinement mechanism and employs a budget-constrained greedy algorithm to strategically select high-value LLM queries, ensuring theoretically grounded, cost-effective optimization. Extensive experiments across eight benchmark datasets demonstrate that Alper significantly outperforms existing cascaded approaches, achieving higher accuracy and robustness under limited query budgets.

Abstract

Dirty entity resolution (ER), which identifies records referring to the same real-world entity from a single, messy dataset, is a fundamental task in data management and mining. However, the dominant blocking-matching-clustering paradigm for ER suffers from critical flaws. Its cascaded, decoupled workflow essentially produces a static, sparse graph plagued by missing edges (due to blocking failures) and noisy links (due to matching errors), causing error propagation and yielding suboptimal clusters, particularly when rigid transitivity is imposed in the clustering. We contend that matching and clustering are fundamentally synergistic, both optimizing for the construction of an ideal entity graph. Building upon this insight, we propose Alper, a unified framework that integrates these steps into an iterative probabilistic label propagation process over a global, evolving graph. Unlike disjoint blocking, Alper refines the graph structure and labels dynamically by adaptively integrating "weak but cheap" signals from graph propagation with "strong but expensive" LLM-based pairwise queries. For higher cost-effectiveness, we formulate the signal selection as a constrained optimization problem maximizing cumulative marginal gain under a query budget, solved via our greedy algorithm with provable theoretical guarantees. Our extensive experiments over eight benchmark datasets demonstrate that Alper is consistently superior to state-of-the-art cascaded pipelines.

Problem

• Entity Resolution
• Dirty Data
• Graph Construction
• Error Propagation
• Clustering

Innovation

• entity resolution
• label propagation
• graph refinement
• large language models
• cost-effective optimization