BEACON: Budget-Aware Entity Matching Across Domains (Extended Technical Report)

2026-03-11

AI Summary

This work addresses the challenge of cross-domain entity matching, where labeled data in the target domain is scarce and expensive to obtain. To tackle this issue, the authors propose the BEACON framework, which leverages pre-trained language models to derive embedding representations of candidate matching pairs and incorporates distribution-aware and budget-aware mechanisms to efficiently select the most transferable source-domain samples for cross-domain training under limited target-label budgets. Evaluated on multiple cross-domain datasets constructed from standard entity matching benchmarks, BEACON significantly outperforms state-of-the-art methods, demonstrating superior robustness and effectiveness particularly in low-resource settings and under varying training budget constraints.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Application Domains: Humanities & Computational Social Science

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Entity Matching (EM)--the task of determining whether two data records refer to the same real-world entity--is a core task in data integration. Recent advances in deep learning have set a new standard for EM, particularly through fine-tuning Pretrained Language Models (PLMs) and, more recently, Large Language Models (LLMs). However, fine-tuning typically requires large amounts of labeled data, which are expensive and time-consuming to obtain. In the context of e-commerce matching, labeling scarcity varies widely across domains, raising the question of how to intelligently train accurate domain-specific EM models with limited labeled data. In this work we assume users have only limited amount of labels for a specific target domain but have access to labeled data from other domains. We introduce BEACON, a distribution-aware, budget-aware framework for low-resource EM across domains. BEACON leverages the insight that embedding representations of pairwise candidate matches can guide the effective selection of out-of-domain samples under limited in-domain supervision. We conduct extensive experiments across multiple domain-partitioned datasets derived from established EM benchmarks, demonstrating that BEACON consistently outperforms state-of-the-art methods under different training budgets.

Problem

• Entity Matching
• Low-Resource Learning
• Cross-Domain
• Label Scarcity
• Data Integration

Innovation

• Entity Matching
• Cross-Domain Learning
• Budget-Aware Learning
• Low-Resource NLP
• Pretrained Language Models