Hierarchical Text Classification on WOS (test)
87.71Micro-F1UME
Evaluation results
| Method | Links | ||
|---|---|---|---|
| UMEEvaluation Protocol=supervised2026.03 | 87.71 | 82.29 | |
| HILLText Encoder=BERT, Learning Paradigm=Contrastive2024.03 | 87.28 | 81.77 | |
| HiTINText Encoder=BERT2023.05 | 87.19 | 81.57 | |
| HGCLRText Encoder=BERT, Learning Paradigm=Contrastive2024.03 | 87.11 | 81.2 | |
| HGCLRText Encoder=BERT2023.05 | 87.11 | 81.2 | |
| BERT+HiMatchText Encoder=BERT, Learning Paradigm=Supervised2024.03 | 86.7 | 81.06 | |
| HiMatchText Encoder=BERT2023.05 | 86.7 | 81.06 | |
| BERT+HTCInfoMaxText Encoder=BERT, Learning Paradigm=Supervised2024.03 | 86.3 | 79.97 | |
| HTCInfoMaxText Encoder=BERT2023.05 | 86.3 | 79.97 | |
| BERT (Chen et al., 2021)Text Encoder=BERT, Learning Paradigm=Supervised2024.03 | 86.26 | 80.58 | |
| HiMatchText Encoder=TextRCNN, Learning Paradigm=Supervised2024.03 | 86.2 | 80.53 | |
| BERT+HiAGMText Encoder=BERT, Learning Paradigm=Supervised2024.03 | 86.04 | 80.19 | |
| HiAGMText Encoder=BERT2023.05 | 86.04 | 80.19 | |
| HiAGMText Encoder=TextRCNN, Learning Paradigm=Supervised2024.03 | 85.82 | 80.28 | |
| BERTText Encoder=BERT, Learning Paradigm=Supervised2024.03 | 85.63 | 79.07 | |
| BERTText Encoder=BERT2023.05 | 85.63 | 79.07 | |
| HTCInfoMaxText Encoder=TextRCNN, Learning Paradigm=Supervised2024.03 | 85.58 | 80.05 | |
| TextRCNNText Encoder=TextRCNN, Learning Paradigm=Supervised2024.03 | 83.55 | 76.99 | |
| LlaMa-3Evaluation Protocol=zero-shot2026.03 | 61.91 | 27.33 | |
| GPT-4Evaluation Protocol=zero-shot2026.03 | 59.74 | 37.8 |