File size: 4,584 Bytes
4c52bbf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
# Scorecard

Results on the recommended evaluation set: the human-annotated records of `megan_donovan_eli_lilly` and
`nora_caterpillar`. Scores are computed with `score.py` on masked characters, with entity type ignored.

| Column | Definition |
|---|---|
| Precision, Recall, F1 | PII detection, on the 1,877 records without unresolved spans |
| Not fully hidden | Share of labelled spans with at least one character left visible |
| Anonymization-grade F1 | Labels and indirect identifiers both required, on all 1,962 records (112 indirect identifiers) |
| Indirect identifiers hidden | Share of indirect identifiers fully masked |
| Records with a leak | Share of the 1,488 records with something to hide in which a required span is partly visible |

The configuration column gives the label set or prompt used for each system. Open models were run with the entity
types listed; other configurations may give different results.

## Token classifiers and span models

| System | Configuration | Precision | Recall | F1 | Not fully hidden % | Anonymization-grade F1 | Indirect identifiers hidden % | Records with a leak % |
|---|---|---|---|---|---|---|---|---|
| **PIIX-560m** | default | 96.5 | 97.2 | 96.9 | 4.0 | 96.2 | 4.5 | 13.8 |
| **PIIX-560m** | `recall=4` | 92.1 | 98.5 | 95.2 | 2.5 | 94.7 | 18.8 | 9.2 |
| **PIIX-560m** | `recall=8` | 82.0 | 99.4 | 89.8 | 1.3 | 89.6 | 58.0 | 5.3 |
| Fastino GLiNER2.5-base | 10 benchmark types | 89.3 | 83.4 | 86.2 | 20.6 | 85.9 | 22.3 | 44.8 |
| Fastino GLiNER2-multi | 10 benchmark types | 85.2 | 84.4 | 84.8 | 18.6 | 83.6 | 48.2 | 42.3 |
| Fastino GLiNER2.5-multi | 10 short queries | 79.8 | 84.1 | 81.9 | 20.4 | 82.6 | 17.0 | 41.4 |
| Fastino GLiNER2-large | 10 benchmark types | 78.5 | 85.5 | 81.8 | 12.6 | 81.6 | 42.9 | 29.1 |
| Perplexity PII-Tracer | native labels | 86.1 | 72.9 | 79.0 | 26.7 | 78.2 | 8.9 | 56.5 |
| GLiNER multi-PII | 53 published types | 72.8 | 85.8 | 78.7 | 16.7 | 79.7 | 8.0 | 36.0 |
| Fastino GLiNER2.5-multi | 10 benchmark types | 85.7 | 69.3 | 76.6 | 41.2 | 77.7 | 9.8 | 61.2 |
| OpenAI Privacy Filter | native labels | 87.6 | 64.1 | 74.1 | 33.7 | 72.2 | 17.0 | 62.6 |
| Fastino GLiNER2 privacy-filter PII multi | 42 native labels | 70.6 | 75.6 | 73.0 | 23.1 | 73.0 | 42.9 | 45.0 |
| Fastino GLiNER2 Guardrails PII Multi | 42 native labels | 67.2 | 77.6 | 72.0 | 21.5 | 71.9 | 52.7 | 41.2 |
| NuMind NuNER Zero | 16 short queries | 61.4 | 79.7 | 69.4 | 25.1 | 71.0 | 31.2 | 22.2 |
| OpenMed privacy-filter-nemotron | 55 native labels | 66.8 | 72.0 | 69.3 | 30.4 | 68.9 | 1.8 | 59.5 |
| GLiNER large v2.1 | 16 short queries | 59.7 | 81.6 | 68.9 | 21.5 | 70.2 | 38.4 | 51.1 |
| Liquid LFM2.5 Encoder 350M PII | 40 native labels | 91.6 | 54.3 | 68.2 | 58.6 | 67.8 | 1.8 | 71.6 |
| PIIBench DeBERTa base | 82 native labels | 53.1 | 94.5 | 68.0 | 10.4 | 68.7 | 9.8 | 27.1 |
| Knowledgator GLiNER PII-large v1.0 | 40 published types | 77.3 | 56.7 | 65.4 | 48.3 | 63.1 | 28.6 | 53.4 |
| Fastino GLiNER2.5-multi | 28 ai4privacy types | 77.4 | 56.3 | 65.2 | 56.3 | 64.7 | 65.2 | 73.3 |
| Microsoft Presidio 2.2.364 | English, spaCy en_core_web_lg | 52.6 | 83.7 | 64.6 | 16.3 | 65.8 | 43.8 | 44.8 |
| Fastino GLiNER2.5-base | 28 ai4privacy types | 68.3 | 60.5 | 64.1 | 48.3 | 63.9 | 68.8 | 82.5 |
| Davlan XLM-R base NER HRL | 4 native labels | 74.8 | 56.1 | 64.1 | 36.3 | 66.2 | 1.8 | 77.8 |
| NVIDIA GLiNER-PII | 55 native labels | 52.1 | 79.3 | 62.9 | 21.1 | 62.4 | 27.7 | 37.7 |
| Knowledgator GLiNER bi-large v1.0 | 16 short queries | 46.5 | 95.6 | 62.6 | 10.7 | 64.6 | 93.8 | 33.8 |
| GLiNER multi v2.1 | 16 short queries | 60.5 | 64.1 | 62.2 | 31.0 | 62.6 | 9.8 | 58.9 |
| Piiranha | 17 native labels | 75.4 | 49.8 | 60.0 | 59.5 | 61.2 | 1.8 | 69.4 |
| Fastino GLiNER2-multi | 28 ai4privacy types | 53.6 | 67.3 | 59.7 | 37.8 | 59.5 | 27.7 | 55.3 |
| Fastino GLiNER2-large | 28 ai4privacy types | 51.1 | 69.7 | 59.0 | 29.1 | 58.8 | 50.9 | 48.6 |
| Gretel GLiNER bi-large v1.0 | 42 published types | 88.3 | 37.3 | 52.4 | 67.2 | 50.2 | 3.6 | 71.9 |
| SpanMarker mBERT MultiNERD | 15 native labels | 64.2 | 39.7 | 49.0 | 54.1 | 51.7 | 1.8 | 89.7 |

## Large language models

| System | Configuration | Precision | Recall | F1 | Not fully hidden % | Anonymization-grade F1 | Indirect identifiers hidden % | Records with a leak % |
|---|---|---|---|---|---|---|---|---|
| GPT-6 Astra | prompted | 99.4 | 96.9 | 98.1 | 6.4 | 97.8 | 0.9 | 22.2 |
| GPT-6.1 Sol | prompted | 99.3 | 96.8 | 98.1 | 6.6 | 97.7 | 0.9 | 22.4 |
| DeepSeek V4.1 Flash | span-extraction prompt | 79.3 | 93.4 | 85.7 | 9.0 | 85.6 | 95.5 | 22.2 |