Named Entity Recognition on multiNERD
96.1Entity F1KnowCoder
Evaluation results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| KnowCoderSetting=Supervised, Model Scale=7B2024.03 | 96.1 | — | — | — | — | — | — | — | — | — | — | — | |
| KnowCoderParameters=7B2026.04 | 96.1 | — | — | — | — | — | — | — | — | — | — | — | |
| KnowCoder-XEvaluation Protocol=Supervised2024.11 | 95.94 | — | — | — | — | — | — | — | — | — | — | — | |
| ProUIEParameters=4B2026.04 | 95.91 | — | — | — | — | — | — | — | — | — | — | — | |
| IEPILEEvaluation Protocol=Supervised2024.11 | 94.6 | — | — | — | — | — | — | — | — | — | — | — | |
| GNEREvaluation Protocol=Supervised2026.04 | 94.4 | — | — | — | — | — | — | — | — | — | — | — | |
| B2NEREvaluation Protocol=Supervised2026.04 | 94 | — | — | — | — | — | — | — | — | — | — | — | |
| B2NEREvaluation Protocol=Supervised2024.11 | 93.98 | — | — | — | — | — | — | — | — | — | — | — | |
| UniNERSetting=Supervised2024.03 | 93.9 | — | — | — | — | — | — | — | — | — | — | — | |
| GLiNEREvaluation Protocol=Supervised2026.04 | 93.8 | — | — | — | — | — | — | — | — | — | — | — | |
| UniversalNERParameters=7B2026.04 | 93.73 | — | — | — | — | — | — | — | — | — | — | — | |
| UniNER-7BEvaluation Protocol=Supervised2026.04 | 93.7 | — | — | — | — | — | — | — | — | — | — | — | |
| KnowCoder-7BEvaluation Protocol=Supervised2026.04 | 93.1 | — | — | — | — | — | — | — | — | — | — | — | |
| InstructUIEBackbone=11B FlanT52023.04 | 92.32 | — | — | — | — | — | — | — | — | — | — | — | |
| BERT-baseEvaluation Protocol=Supervised2026.04 | 91.3 | — | — | — | — | — | — | — | — | — | — | — | |
| BERTEvaluation Protocol=Supervised2024.11 | 91.25 | — | — | — | — | — | — | — | — | — | — | — | |
| Bert-baseBackbone=Bert-base2023.04 | 91.25 | — | — | — | — | — | — | — | — | — | — | — | |
| BERT-base2026.04 | 91.25 | — | — | — | — | — | — | — | — | — | — | — | |
| InstructUIEEvaluation Protocol=Supervised2026.04 | 90.3 | — | — | — | — | — | — | — | — | — | — | — | |
| InstructUIEParameters=11B2026.04 | 90.26 | — | — | — | — | — | — | — | — | — | — | — | |
| YAYI-UIEEvaluation Protocol=Supervised2024.11 | 88.42 | — | — | — | — | — | — | — | — | — | — | — | |
| InstructUIE2023.04 | 85 | — | — | — | — | — | — | — | — | — | — | — | |
| GLADEnsemble=true2026.04 | 81.8 | — | — | — | — | — | — | — | — | — | — | — | |
| MVEnsemble=true2026.04 | 81.5 | — | — | — | — | — | — | — | — | — | — | — | |
| MACEEnsemble=true2026.04 | 81.5 | — | — | — | — | — | — | — | — | — | — | — | |
| DiZiNERZero-shot=true2026.04 | 80.6 | — | — | — | — | — | — | — | — | — | — | — | |
| DiZiNEREvaluation Protocol=Zero-shot2026.04 | 80.6 | — | — | — | — | — | — | — | — | — | — | — | |
| GoLLIEModel size=13B, Zero-shot=true2023.10 | 77.5 | — | — | — | — | — | — | — | — | — | — | — | |
| Prior Best ZSZero-shot=true2026.04 | 77.5 | — | — | — | — | — | — | — | — | — | — | — | |
| GoLLIEEvaluation Protocol=Zero-shot2026.04 | 77.5 | — | — | — | — | — | — | — | — | — | — | — | |
| GoLLIEModel size=34B, Zero-shot=true2023.10 | 77.2 | — | — | — | — | — | — | — | — | — | — | — | |
| GoLLIEModel size=7B, Zero-shot=true2023.10 | 76 | — | — | — | — | — | — | — | — | — | — | — | |
| GPT-5 miniZero-shot=true2026.04 | 74 | — | — | — | — | — | — | — | — | — | — | — | |
| GPT-5 miniEvaluation Protocol=Zero-shot, Note=supervisor model2026.04 | 74 | — | — | — | — | — | — | — | — | — | — | — | |
| LOCUSsetup=few-shot2025.12 | 70.77 | — | 71.66 | 70.98 | 69.99 | 69.77 | 69.82 | 71.43 | 72.18 | — | — | — | |
| LOCUSminisetup=few-shot2025.12 | 69.33 | — | 71.84 | 71.61 | 68.32 | 70.62 | 66.21 | 70.17 | 72.33 | — | — | — | |
| GPT-4osetup=few-shot2025.12 | 68.55 | — | 70.19 | 70.34 | 70.09 | 67.17 | 66.06 | 65.93 | 70.07 | — | — | — | |
| DSEnsemble=true2026.04 | 68.3 | — | — | — | — | — | — | — | — | — | — | — | |
| JPT-4Bzero-shot=true2026.04 | 65.7 | — | — | — | — | — | — | — | — | — | — | — | |
| GLiNER-Lzero-shot=true2026.04 | 59.7 | — | — | — | — | — | — | — | — | — | — | — | |
| GLiNEREvaluation Protocol=Zero-shot2026.04 | 59.7 | — | — | — | — | — | — | — | — | — | — | — | |
| gliner_medium-v2.5Evaluation Protocol=Zero-shot2026.02 | 59.6 | — | — | — | — | — | — | — | — | — | — | — | |
| UniNER-7Bzero-shot=true2026.04 | 59.3 | — | — | — | — | — | — | — | — | — | — | — | |
| UniNER-7BEvaluation Protocol=Zero-shot2026.04 | 59.3 | — | — | — | — | — | — | — | — | — | — | — | |
| ChatGPTEvaluation Protocol=Zero-shot2026.04 | 58.1 | — | — | — | — | — | — | — | — | — | — | — | |
| gliner_small-v2.5Evaluation Protocol=Zero-shot2026.02 | 57 | — | — | — | — | — | — | — | — | — | — | — | |
| GoLLIEModel variant=Baseline, Zero-shot=true2023.10 | 55 | — | — | — | — | — | — | — | — | — | — | — | |
| gliner_large-v2.5Evaluation Protocol=Zero-shot2026.02 | 50.9 | — | — | — | — | — | — | — | — | — | — | — | |
| ToMMeRDecoding Strategy=Greedy (flat) decoding, Zero-shot=true, Backbone=LLAMA3.2-1B, Plugged Layer=6, Parameters=274K, Nested=false, #samples=154 144, #entities=238,0052025.10 | 45.5 | — | — | — | — | — | — | — | — | — | 94 | 30 | |
| ToMMeRDecoding Strategy=Threshold decoding, Zero-shot=true, Backbone=LLAMA3.2-1B, Plugged Layer=6, Parameters=274K, Nested=false, #samples=154 144, #entities=238,0052025.10 | 35.5 | — | — | — | — | — | — | — | — | — | 98.6 | 21.7 | |
| Gemma3-27B2026.01 | — | 64.6 | — | — | — | — | — | — | — | — | — | — | |
| GLiNERBackbone=deBERTa-v3, #Trained Params=209M2025.10 | — | — | — | — | — | — | — | — | — | 93.8 | — | — | |
| GLiNER-multi-v2.12026.01 | — | 53.3 | — | — | — | — | — | — | — | — | — | — | |
| GLiNER-x-base2026.01 | — | 58.2 | — | — | — | — | — | — | — | — | — | — | |
| GPT-52026.01 | — | 49.6 | — | — | — | — | — | — | — | — | — | — | |
| OTTERArchitecture=Bi-Encoder, Backbone=RemBERT2026.01 | — | 59.5 | — | — | — | — | — | — | — | — | — | — | |
| OTTERArchitecture=Bi-Encoder, Backbone=mmBERT2026.01 | — | 56.9 | — | — | — | — | — | — | — | — | — | — | |
| OTTERArchitecture=Bi-Encoder, Backbone=mmBERT, Training Steps=100k2026.01 | — | 54.4 | — | — | — | — | — | — | — | — | — | — | |
| OTTERArchitecture=Cross-Encoder, Backbone=RemBERT2026.01 | — | 63.6 | — | — | — | — | — | — | — | — | — | — | |
| OTTERArchitecture=Cross-Encoder, Backbone=mmBERT2026.01 | — | 62.7 | — | — | — | — | — | — | — | — | — | — | |
| OTTERArchitecture=Cross-Encoder, Backbone=mmBERT, Training Steps=100k2026.01 | — | 63.8 | — | — | — | — | — | — | — | — | — | — | |
| OTTERArchitecture=Cross-Encoder, Backbone=mmBERT, Training Steps=100k, Thresholding Strategy=Best per language2026.01 | — | 66.1 | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-32B2026.01 | — | 61.7 | — | — | — | — | — | — | — | — | — | — | |
| ToMMeR + span embed (ℓ11)Backbone=LLaMA-3.2-1B, #Trained Params=7.6M, Probed layer=ℓ62025.10 | — | — | — | — | — | — | — | — | — | 92.2 | — | — | |
| ToMMeR + span embed (ℓ11)Backbone=LLaMA-3.2-3B, #Trained Params=7.7M, Probed layer=ℓ52025.10 | — | — | — | — | — | — | — | — | — | 93.3 | — | — | |
| ToMMeR + span embed (ℓ11)Backbone=LLaMA-3.1-8B, #Trained Params=7.9M, Probed layer=ℓ52025.10 | — | — | — | — | — | — | — | — | — | 92.4 | — | — | |
| ToMMeR + span embed (ℓ11)Backbone=RoBERTa-base, #Trained Params=7.4M, Probed layer=ℓ52025.10 | — | — | — | — | — | — | — | — | — | 92.6 | — | — | |
| ToMMeR + span embed (ℓ11)Backbone=BERT-base, #Trained Params=7.4M, Probed layer=ℓ32025.10 | — | — | — | — | — | — | — | — | — | 90.4 | — | — | |
| WikiNeural2026.01 | — | 65.2 | — | — | — | — | — | — | — | — | — | — |