Relation Extraction on BioRED
34.31Overall Rel+ F1HGNet
Evaluation results
| Method | Links | |
|---|---|---|
| HGNetLearning Paradigm=Supervised, Loss Configuration=w/o LDHL2026.03 | 34.31 | |
| HGNetLearning Paradigm=Supervised2026.03 | 33.85 | |
| HGNetLearning Paradigm=Supervised, Loss Configuration=w/o LCAF Loss2026.03 | 33.09 | |
| GATLearning Paradigm=Supervised, Loss Configuration=w/o LDHL2026.03 | 32.74 | |
| GATLearning Paradigm=Supervised2026.03 | 32.4 | |
| HGERELearning Paradigm=Supervised2026.03 | 32.39 | |
| GCNLearning Paradigm=Supervised, Loss Configuration=w/o LDHL2026.03 | 32.28 | |
| GCNLearning Paradigm=Supervised2026.03 | 31.93 | |
| PL-MarkerLearning Paradigm=Supervised2026.03 | 29.87 | |
| PURELearning Paradigm=Supervised2026.03 | 29.41 | |
| llama-3.3-70b-versatileLearning Paradigm=Zero-Shot2026.03 | 7.29 | |
| openai/gpt-oss-120bLearning Paradigm=Zero-Shot2026.03 | 7.15 | |
| qwen/qwen3-32bLearning Paradigm=Zero-Shot2026.03 | 6.71 | |
| GPT-3.5 TurboLearning Paradigm=Zero-Shot2026.03 | 6.36 | |
| llama-3.1-8b-instantLearning Paradigm=Zero-Shot2026.03 | 5.48 |