EN · PII
kiji-en PII detection benchmark results
A frozen evaluation slice with 1,000 rows, 7,623 normalized gold annotations and source-level provenance.
1,000input rows
7,623gold annotations
385,103characters
41measured detectors
Dataset provenance
- Language group
- EN
- Task
- PII
- License
- apache-2.0
- Data categories
- ACCOUNT, ADDRESS, CONTACT, DATE, ID, NET, ORG, 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 fully hidden annotations. Training-source overlaps are excluded.
| Detector | Family | Fully hidden ↑ | Detected / any overlap ↑ | Extra masking ↓ | Char F1 ↑ |
|---|---|---|---|---|---|
| nym-small | ONNX | 96.09% | 97.74% | — | 0.907 |
| gliner2-fastino | GLINER2 | 96.05% | 96.79% | — | 0.878 |
| pplx | PPLX | 95.74% | 95.93% | — | 0.863 |
| nym-base | HF | 95.36% | 97.26% | — | 0.901 |
| gravitee-small | HF | 95.26% | 95.75% | — | 0.923 |
| gliner2-hivetrace-omni | GLINER2 | 95.04% | 95.64% | — | 0.883 |
| openmed-multilingual | HF | 93.31% | 95.17% | — | 0.894 |
| gliner-nvidia | GLINER | 93.11% | 96.37% | — | 0.874 |
| bardsai-eu | ONNX | 89.69% | 95.82% | — | 0.842 |
| openmed-nemotron | HF | 89.09% | 91.77% | — | 0.856 |
| kalyan-ettin | HF | 86.00% | 91.07% | — | 0.845 |
| gliner2-large | GLINER2 | 85.92% | 86.72% | — | 0.850 |
| gliner-stream-pii | GLINER | 85.82% | 90.06% | — | 0.849 |
| apararti | OPF | 84.77% | 85.39% | — | 0.825 |
| openai-base | OPF | 84.36% | 84.81% | — | 0.810 |
| opf-kz-ru | OPF | 82.82% | 83.75% | — | 0.824 |
| mmbert32k | HF | 77.45% | 87.16% | — | 0.845 |
| traciora | OPF | 76.64% | 79.92% | — | 0.794 |
| gliner25-fastino | GLINER2 | 74.09% | 75.00% | — | 0.802 |
| nuner-zero | GLINER | 73.68% | 76.16% | — | 0.804 |
| opf-ru-v2 | OPF | 73.61% | 76.69% | — | 0.788 |
| gliner2-hivetrace-uni | GLINER2 | 71.43% | 72.44% | — | 0.788 |
| stanza-ru | STANZA | 70.38% | 72.64% | — | 0.458 |
| opf-ru | OPF | 69.15% | 75.30% | — | 0.783 |
| gliner-multi-v21 | GLINER | 67.32% | 69.70% | — | 0.668 |
| ru-pii-ner | RUPII | 66.84% | 67.60% | — | 0.715 |
| gliner-pii-base | GLINER | 62.06% | 63.64% | — | 0.750 |
| gliner-urchade | GLINER | 60.13% | 61.59% | — | 0.746 |
| davlan-xlmr | HF | 60.04% | 62.44% | — | 0.646 |
| gliner-pii-edge | GLINER | 59.11% | 60.59% | — | 0.718 |
| davlan-mbert | HF | 56.33% | 57.96% | — | 0.630 |
| ner-ru-gherman | HF | 56.25% | 61.56% | — | 0.591 |
| gliner2-vladlinv | GLINER2 | 54.87% | 55.21% | — | 0.663 |
| ner-ru-yqelz | HF | 51.65% | 55.06% | — | 0.575 |
| pii-shield-onnx | ONNX | 37.58% | 43.46% | — | 0.637 |
| ru-legal-ner | HF | 28.02% | 36.76% | — | 0.554 |
| spacy-ru-lg | SPACY | 20.54% | 20.66% | — | 0.243 |
| natasha | NATASHA | 14.04% | 14.19% | — | 0.203 |
| fef2-secret-ru | HF | 13.71% | 15.18% | — | 0.310 |
| rules-ru | RULES | 7.84% | 8.13% | — | 0.317 |
| spacy-alrosait | SPACY | 0.00% | 0.00% | — | 0.000 |