Papers › WDC Products: A Multi-Dimensional Entity Matching Benchmark

WDC Products: A Multi-Dimensional Entity Matching Benchmark

23 Jan 2023arXiv:2301.09521archive 2025-07-28

Ralph Peeters, Reng Chiz Der, Christian Bizer

The difficulty of an entity matching task depends on a combination of multiple factors such as the amount of corner-case pairs, the fraction of entities in the test set that have not been seen during training, and the size of the development set. Current entity matching benchmarks usually represent single points in the space along such dimensions or they provide for the evaluation of matching methods along a single dimension, for instance the amount of training data. This paper presents WDC Products, an entity matching benchmark which provides for the systematic evaluation of matching systems along combinations of three dimensions while relying on real-world data. The three dimensions are (i) amount of corner-cases (ii) generalization to unseen entities, and (iii) development set size (training set plus validation set). Generalization to unseen entities is a dimension not covered by any of the existing English-language benchmarks yet but is crucial for evaluating the robustness of entity matching systems. Instead of learning how to match entity pairs, entity matching can also be formulated as a multi-class classification task that requires the matcher to recognize individual entities. WDC Products is the first benchmark that provides a pair-wise and a multi-class formulation of the same tasks. We evaluate WDC Products using several state-of-the-art matching systems, including Ditto, HierGAT, and R-SupCon. The evaluation shows that all matching systems struggle with unseen entities to varying degrees. It also shows that for entity matching contrastive learning is more training data efficient compared to cross-encoders.

PaperPDFCode

Code

wbsg-uni-mannheim/wdcproducts officialmentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Contrastive LearningData IntegrationEntity ResolutionMulti-class Classification

Datasets

Introduced by this paper, per the archive.

WDC Products

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Entity Resolution WDC Products-50%cc-unseen-medium RoBERTa-base F1 (%) 71.14 #1 of 4 Archive leaderboard report
Entity Resolution WDC Products-50%cc-unseen-medium Ditto F1 (%) 70.66 #2 of 4 Archive leaderboard report
Entity Resolution WDC Products-50%cc-unseen-medium HG F1 (%) 68.74 #3 of 4 Archive leaderboard report
Entity Resolution WDC Products-50%cc-unseen-medium RoBERTa-SupCon F1 (%) 57.23 #4 of 4 Archive leaderboard report
Entity Resolution WDC Products-80%cc-seen-medium RoBERTa-SupCon F1 (%) 79.99 #5 of 13 Archive leaderboard report
Entity Resolution WDC Products-80%cc-seen-medium Ditto F1 (%) 73.93 #10 of 13 Archive leaderboard report
Entity Resolution WDC Products-80%cc-seen-medium RoBERTa-base F1 (%) 72.18 #11 of 13 Archive leaderboard report
Entity Resolution WDC Products-80%cc-seen-medium HG F1 (%) 71.40 #12 of 13 Archive leaderboard report
Entity Resolution WDC Products-80%cc-seen-medium-multi RoBERTa-SupCon F1 Micro 88.63 #1 of 2 Archive leaderboard report
Entity Resolution WDC Products-80%cc-seen-medium-multi RoBERTa-base F1 Micro 52.03 #2 of 2 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Contrastive LearningTest

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections