Citation prediction on Paecter citation dataset
1.26RFRTechToken BERT
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TechToken BERTFine-tuned=on the Paecter citation dataset, Selection Strategy=Best across mean-pool and CLS embedding2026.05 | 1.26 | 68.96 | 89.54 | |
| PaecterSelection Strategy=Best across mean-pool and CLS embedding2026.05 | 1.32 | 68.11 | 88.14 | |
| BERT4PatentsSelection Strategy=Best across mean-pool and CLS embedding2026.05 | 1.57 | 59.46 | 80.6 | |
| TechToken BERTSelection Strategy=Best across mean-pool and CLS embedding2026.05 | 1.59 | 59.73 | 79.7 | |
| LLaMA PatentsSelection Strategy=Best across mean-pool and CLS embedding2026.05 | 1.74 | 56.78 | 75.47 | |
| PatentSBERTaSelection Strategy=Best across mean-pool and CLS embedding2026.05 | 1.79 | 55.89 | 75.95 | |
| BERT4Patents FTSelection Strategy=Best across mean-pool and CLS embedding2026.05 | 1.88 | 52.78 | 73.16 |