Named Entity Recognition on CrossNER + MIT Aggregate
74.1Average ScoreJPT-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| JPT-8Bevaluation_mode=zero-shot, paradigm=Discriminative2026.04 | 74.1 | |
| JPT-4Bevaluation_mode=zero-shot, paradigm=Discriminative2026.04 | 70.9 | |
| SaMBase Model=Llama3.1-8B, Training Strategy=SaM, Evaluation Setting=Zero-shot2025.06 | 66.24 | |
| SaM (MoE)evaluation_mode=zero-shot, paradigm=Generative2026.04 | 66.2 | |
| B2NEREvaluation Setting=Zero-shot2025.06 | 64.57 | |
| SaMBase Model=Qwen2.5-7B, Training Strategy=SaM, Evaluation Setting=Zero-shot2025.06 | 63.13 | |
| UniNER-7Bevaluation_mode=zero-shot, paradigm=Generative2026.04 | 61.8 | |
| UniNEREvaluation Setting=Zero-shot2025.06 | 61.79 | |
| LlamaBase Model=Llama3.1-8B, Training Strategy=Fully-trained, Evaluation Setting=Zero-shot2025.06 | 60.94 | |
| GLiNEREvaluation Setting=Zero-shot, Parameters=300M2025.06 | 60.9 | |
| GLiNER-Levaluation_mode=zero-shot, paradigm=Discriminative2026.04 | 60.9 | |
| KnowCoderEvaluation Setting=Zero-shot2025.06 | 60.13 | |
| GoLLIEevaluation_mode=zero-shot, paradigm=Generative2026.04 | 58.4 | |
| GoLLIEEvaluation Setting=Zero-shot2025.06 | 58.39 | |
| QwenBase Model=Qwen2.5-7B, Training Strategy=Fully-trained, Evaluation Setting=Zero-shot2025.06 | 57.1 | |
| InstructUIEEvaluation Setting=Zero-shot2025.06 | 47.84 | |
| InstructUIEevaluation_mode=zero-shot, paradigm=Generative2026.04 | 47.8 |