EN · PII
nym-en PII detection benchmark results
A frozen evaluation slice with 734 rows, 3,584 normalized gold annotations and source-level provenance.
734input rows
3,584gold annotations
215,898characters
39measured detectors
Dataset provenance
- Language group
- EN
- Task
- PII
- License
- mit
- Data categories
- ACCOUNT, ADDRESS, CONTACT, DATE, ID, NET, ORG, OTHER, PERSON, SECRET
- Project generated
- No
- Corrupted copy
- No
Rows without annotations are not verified clean. Reported extra masking is not a human-confirmed false-positive rate.
Detector outcomes
Sorted by untouched annotations. Training-source overlaps are excluded.
| Detector | Family | Untouched ↓ | Fully hidden ↑ | Extra masking ↓ | Char F1 ↑ |
|---|---|---|---|---|---|
| pplx | PPLX | 3.85% | 94.89% | 2.15% | 0.927 |
| openmed-multilingual | HF | 4.10% | 91.29% | 5.25% | 0.925 |
| mmbert32k | HF | 8.59% | 77.57% | 4.02% | 0.885 |
| gliner-nvidia | GLINER | 11.27% | 86.19% | 9.92% | 0.892 |
| kalyan-ettin | HF | 12.56% | 76.93% | 2.02% | 0.857 |
| openmed-nemotron | HF | 13.59% | 77.29% | 2.28% | 0.844 |
| gravitee-small | HF | 13.76% | 84.77% | 3.80% | 0.862 |
| apararti | OPF | 14.62% | 82.06% | 2.47% | 0.869 |
| opf-kz-ru | OPF | 14.87% | 81.64% | 2.81% | 0.857 |
| gliner2-fastino | GLINER2 | 15.63% | 81.89% | 1.68% | 0.876 |
| bardsai-eu | ONNX | 16.96% | 73.47% | 0.39% | 0.814 |
| openai-base | OPF | 18.22% | 78.43% | 1.91% | 0.855 |
| ru-pii-ner | RUPII | 21.07% | 77.04% | 1.84% | 0.831 |
| gliner2-hivetrace-omni | GLINER2 | 21.34% | 75.81% | 0.50% | 0.849 |
| opf-ru | OPF | 22.57% | 65.71% | 4.45% | 0.811 |
| gliner-stream-pii | GLINER | 25.81% | 69.14% | 7.96% | 0.790 |
| nuner-zero | GLINER | 29.10% | 68.61% | 3.46% | 0.811 |
| gliner2-large | GLINER2 | 29.55% | 66.96% | 0.54% | 0.776 |
| gliner-pii-edge | GLINER | 32.09% | 65.96% | 9.50% | 0.764 |
| gliner2-hivetrace-uni | GLINER2 | 36.38% | 61.38% | 4.42% | 0.784 |
| opf-ru-v2 | OPF | 36.83% | 56.61% | 0.31% | 0.747 |
| gliner2-vladlinv | GLINER2 | 38.81% | 60.18% | 1.13% | 0.731 |
| traciora | OPF | 40.88% | 51.95% | 1.34% | 0.715 |
| pii-shield-onnx | ONNX | 47.94% | 39.23% | 2.69% | 0.629 |
| gliner25-fastino | GLINER2 | 48.52% | 49.30% | 0.63% | 0.680 |
| stanza-ru | STANZA | 48.60% | 43.78% | 9.41% | 0.358 |
| gliner-pii-base | GLINER | 49.39% | 50.36% | 2.31% | 0.690 |
| ru-legal-ner | HF | 50.20% | 33.43% | 6.25% | 0.630 |
| gliner-urchade | GLINER | 51.95% | 47.63% | 2.68% | 0.682 |
| davlan-xlmr | HF | 57.51% | 37.89% | 0.16% | 0.473 |
| ner-ru-yqelz | HF | 57.76% | 38.48% | 4.79% | 0.458 |
| davlan-mbert | HF | 58.98% | 35.60% | 0.53% | 0.457 |
| gliner-multi-v21 | GLINER | 59.71% | 38.11% | 0.26% | 0.479 |
| ner-ru-gherman | HF | 62.92% | 29.24% | 0.14% | 0.396 |
| fef2-secret-ru | HF | 87.86% | 11.10% | 0.71% | 0.271 |
| spacy-ru-lg | SPACY | 89.01% | 9.88% | 9.02% | 0.148 |
| rules-ru | RULES | 89.31% | 10.44% | 0.00% | 0.302 |
| natasha | NATASHA | 89.70% | 9.26% | 0.18% | 0.137 |
| spacy-alrosait | SPACY | 100.00% | 0.00% | 0.00% | 0.000 |