Entity Matching on DBLP-Scholar (test)
95.64F1 ScoreDitto (Benchmark set)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Ditto (Benchmark set)Model=Ditto (RoBERTa), Training Set=Gold Benchmark2026.06 | 95.64 | — | — | |
| Ditto (Best m.l. set)Model=Ditto (RoBERTa), Training Set=Machine-labeled (Best strategy)2026.06 | 93.86 | — | — | |
| Kimi K2.6Matching Strategy=Zero-shot direct matching2026.06 | 92.49 | 3.15 | 1.37 | |
| GPT-5.2Matching Strategy=Zero-shot direct matching2026.06 | 90.87 | — | — | |
| Qwen 3.6 PlusMatching Strategy=Zero-shot direct matching2026.06 | 88.96 | — | — |