Entity Matching on Average (Febrl4, F.-Zagat, Bikes, eBooks, Movies)
91.2F1 ScoreFlexTab-Match
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| FlexTab-MatchLearning Paradigm=In-context learners, Validation Protocol=5-fold cross-validation2026.06 | 91.2 | 1 | 2.2 | |
| FlexTab-MatchModel Category=In-context learners2026.06 | 91.2 | 1 | 2.4 | |
| Deepmatch HybridLearning Paradigm=Per-dataset trained, Validation Protocol=5-fold cross-validation2026.06 | 90.2 | 1.6 | 2.2 | |
| Deepmatch HybridModel Category=Per-dataset trained2026.06 | 90.2 | 2 | 2.6 | |
| Deepmatch RNNModel Category=Per-dataset trained2026.06 | 88.8 | 2.2 | 2.8 | |
| Deepmatch AttnModel Category=Per-dataset trained2026.06 | 87.8 | 2.6 | 4 | |
| DITTOLearning Paradigm=Per-dataset trained, Validation Protocol=5-fold cross-validation2026.06 | 84.9 | 1.2 | 2 | |
| DITTOModel Category=Per-dataset trained2026.06 | 84.9 | 1.6 | 2.4 | |
| FlexTab EmbeddingsLearning Paradigm=Embedding similarity, Validation Protocol=5-fold cross-validation2026.06 | 79.6 | 1.8 | 5.4 | |
| FlexTab EmbeddingsModel Category=Embedding similarity2026.06 | 79.6 | 1.8 | 7.8 | |
| NGramHash EmbeddingsLearning Paradigm=Embedding similarity, Validation Protocol=5-fold cross-validation2026.06 | 73.7 | 1.4 | 4.4 | |
| NGramHash EmbeddingsModel Category=Embedding similarity2026.06 | 73.7 | 1.6 | 6.4 | |
| ConTextTabLearning Paradigm=In-context learners, Validation Protocol=5-fold cross-validation2026.06 | 65.6 | 3.6 | 6.8 | |
| ConTextTabModel Category=In-context learners2026.06 | 65.6 | 4.2 | 9.6 | |
| FlexTabLearning Paradigm=In-context learners, Validation Protocol=5-fold cross-validation2026.06 | 62.1 | 3.6 | 7 | |
| FlexTabModel Category=In-context learners2026.06 | 62.1 | 3.8 | 9.4 | |
| TabICLLearning Paradigm=In-context learners, Validation Protocol=5-fold cross-validation2026.06 | 57.5 | 2.8 | 6.8 | |
| TabICLModel Category=In-context learners2026.06 | 57.5 | 3.2 | 9.6 | |
| TabPFNLearning Paradigm=In-context learners, Validation Protocol=5-fold cross-validation2026.06 | 55.3 | 3.2 | 7.2 | |
| TabPFNModel Category=In-context learners2026.06 | 55.3 | 3.6 | 10 | |
| TabPFN (TF-IDF)Model Category=In-context learners, Input Representation=TF-IDF2026.06 | 54.7 | 5 | 11 | |
| TabPFN EmbeddingsLearning Paradigm=Embedding similarity, Validation Protocol=5-fold cross-validation2026.06 | 48.9 | 2.8 | 8.6 | |
| TabPFN EmbeddingsModel Category=Embedding similarity2026.06 | 48.9 | 3.2 | 12.6 | |
| Skrub EmbeddingsModel Category=Embedding similarity2026.06 | 47.7 | 3.4 | 12.6 | |
| TabICL (TF-IDF)Model Category=In-context learners, Input Representation=TF-IDF2026.06 | 47.3 | 5 | 11.4 | |
| Deepmatch SIFModel Category=Per-dataset trained2026.06 | 27 | 4.2 | 12.4 |