Better Retrieval, Limited Clustering Gains: A Controlled Study of Multilingual Company Entity Resolution

2026-10-04

AI Summary

This study investigates whether improvements in retrieval performance effectively translate into clustering quality gains in multilingual corporate entity resolution. Through controlled experiments built upon a multilingual E5 encoder with hard negative training, we systematically analyze the relationship between retrieval recall and downstream clustering. For the first time, this work quantitatively demonstrates that, under a fixed confirmation classifier, substantial increases in retrieval recall can paradoxically degrade clustering precision, as newly retrieved positive pairs predominantly fall below the decision threshold. This finding reveals a bottleneck suppression effect inherent to the confirmation stage, underscoring that evaluation frameworks must account for end-to-end clustering quality rather than retrieval metrics alone. Ultimately, these insights provide critical empirical evidence for optimizing multilingual entity resolution pipelines.

Abstract

Improved name retrieval may have little effect on company clusters when the pair classifier remains unchanged. We examine this dependency by adapting multilingual E5 encoders under fixed candidate budgets and downstream decision rules. Random-negative and hard-negative training use identical positive schedules. Checkpoints are selected before collecting a new GLEIF sample of 3,633 names, 2,880 source identities and 882 silver-positive pairs. At 72,660 candidate edges, adaptation with a multi-view selector increases direct pair recall from 53.74% to 76.98%. The primary matcher adds only seven correct and two incorrect co-cluster pairs: cluster recall rises from 32.54% to 33.33%, while precision falls from 95.99% to 95.45%. Of 208 newly retrieved silver-positive pairs, 202 fall below its decision threshold. Random-negative and hard-negative training produce identical final partitions. An AI-assisted, single-reviewer audit of 137 pairs supports the observed pattern, although its predominantly LEI-derived evidence does not establish independent gold labels. The results locate the immediate loss of retrieval gains at the existing confirmation stage and show why encoder evaluation must also measure final cluster quality.

Problem

• Entity Resolution
• Multilingual Retrieval
• Clustering Quality
• Pair Classification
• Company Name Matching

Innovation

• Entity Resolution
• Multilingual Encoders
• Retrieval-Clustering Gap
• Hard-Negative Training
• Cluster Quality