
nemotron (100000 docs, floor medium)
category           gold     tp     fp          precision             recall         recall valid
aadhaar               0      0     28         0.0 [0-12]                  -              - n=0
account_pin        6456   4999     37      99.3 [99-100]       77.4 [76-78]   77.4 [76-78] n=6456
address           16879  11589    154       98.7 [98-99]       68.7 [68-69]   68.7 [68-69] n=16879
bank_account      16693  13557    400       97.1 [97-97]       81.2 [81-82]   81.2 [81-82] n=16693
bank_routing       8354    783      4      99.5 [99-100]         9.4 [9-10]   92.9 [91-94] n=843
card              12867   1514     13      99.1 [98-100]       11.8 [11-12]  99.8 [99-100] n=1517
coordinates        7659   6591     72       98.9 [99-99]       86.1 [85-87]   86.1 [85-87] n=7659
cvv                4843   4068     25      99.4 [99-100]       84.0 [83-85]   84.0 [83-85] n=4843
date_of_birth     18079  17863    223       98.8 [99-99]       98.8 [99-99]   98.8 [99-99] n=18079
driving_licence       0      0     28         0.0 [0-12]                  -              - n=0
email             53930  53481    117     99.8 [100-100]       99.2 [99-99]  99.7 [100-100] n=53655
gender             7645   2424    257       90.4 [89-92]       31.7 [31-33]   31.7 [31-33] n=7645
health             5539      1   1053          0.1 [0-0]          0.0 [0-0]      0.0 [0-0] n=5539
health_id         21656  19468     40     99.8 [100-100]       89.9 [90-90]   89.9 [90-90] n=21656
iban                  0      0      7         0.0 [0-35]                  -              - n=0
id_document        5849      0      0                  -          0.0 [0-0]      0.0 [0-0] n=5849
ip_address         9217   3395      5     99.9 [100-100]       36.8 [36-38]   42.6 [42-44] n=7975
mac_address        4250   4245      9     99.8 [100-100]     99.9 [100-100]  99.9 [100-100] n=4250
pan                   0      0     15        0.0 [-0-20]                  -              - n=0
passport              0      0     11         0.0 [0-26]                  -              - n=0
person_name      143639  42438    868       98.0 [98-98]       29.5 [29-30]   29.5 [29-30] n=143639
phone             30353  24678   1174       95.5 [95-96]       81.3 [81-82]  99.9 [100-100] n=24666
postal_code        6279   2039    107       95.0 [94-96]       32.5 [31-34]   32.5 [31-34] n=6279
secret            11652   8989   1775       83.5 [83-84]       77.1 [76-78]   77.1 [76-78] n=11652
ssn                6062   6000    807       88.1 [87-89]       99.0 [99-99]  99.4 [99-100] n=6036
swift_bic          5559   3555     16      99.6 [99-100]       64.0 [63-65]   87.3 [86-88] n=4072
tax_id             1302    975     15       98.5 [98-99]       74.9 [72-77]   74.9 [72-77] n=1302
vehicle_reg        4078   3699     43       98.9 [98-99]       90.7 [90-92]   90.7 [90-92] n=4078
vin                4561   3663     44       98.8 [98-99]       80.3 [79-81] 100.0 [99-100] n=287
unmapped: date 73867 (0.3%), company_name 54837 (0.2%), url 37847 (0.4%), occupation 36888 (0.2%), time 24506 (0.0%), country 23475 (0.8%), customer_id 20502 (3.1%), city 18347 (3.0%), state 18318 (2.6%), user_name 15800 (0.7%), date_time 13221 (0.1%), biometric_identifier 11379 (0.2%), employment_status 11018 (0.1%), education_level 9378 (0.0%), race_ethnicity 9248 (0.2%), employee_id 8878 (0.0%), language 8651 (0.1%), county 8465 (0.8%), age 7167 (0.1%), political_view 6402 (0.1%), religious_belief 6257 (0.0%), http_cookie 4892 (28.1%), sexuality 3312 (0.2%), device_identifier 2507 (0.3%), unique_id 1777 (6.8%)

gretel-finance (5594 docs, floor medium)
category           gold     tp     fp          precision             recall         recall valid
aadhaar               0      0     11         0.0 [0-26]                  -              - n=0
account_pin         143     69      1      98.6 [92-100]       48.3 [40-56]   48.3 [40-56] n=143
address            4307   1195    290       80.5 [78-82]       27.7 [26-29]   27.7 [26-29] n=4307
bank_account        166     69    263       20.8 [17-26]       41.6 [34-49]   41.6 [34-49] n=166
bank_routing        158      3      2       60.0 [23-88]          1.9 [1-5] 100.0 [44-100] n=3
card                120     49      8       86.0 [75-93]       40.8 [32-50] 100.0 [93-100] n=49
coordinates          97     63     47       57.3 [48-66]       64.9 [55-74]   64.9 [55-74] n=97
crypto_wallet         0      0     83          0.0 [0-4]                  -              - n=0
cvv                 160     65      4       94.2 [86-98]       40.6 [33-48]   40.6 [33-48] n=160
date_of_birth       250    146     38       79.3 [73-85]       58.4 [52-64]   58.4 [52-64] n=250
driving_licence     140     55     10       84.6 [74-91]       39.3 [32-48]   39.3 [32-48] n=140
email              1422   1347     45       96.8 [96-98]       94.7 [93-96]   97.5 [97-98] n=1381
gender                0      0      7         0.0 [0-35]                  -              - n=0
health                0      0      8         0.0 [0-32]                  -              - n=0
health_id             0      0      4         0.0 [0-49]                  -              - n=0
iban                203    156      4       97.5 [94-99]       76.8 [71-82]   95.6 [91-98] n=158
ip_address          298     73      0     100.0 [95-100]       24.5 [20-30]   29.4 [24-35] n=248
passport            136     58      1      98.3 [91-100]       42.6 [35-51]   42.6 [35-51] n=136
person_name       10812    950     48       95.2 [94-96]          8.8 [8-9]      8.8 [8-9] n=10812
phone               946    402    101       79.9 [76-83]       42.5 [39-46]   93.2 [90-95] n=427
postal_code           0      0    149          0.0 [0-2]                  -              - n=0
secret              192     83     20       80.6 [72-87]       43.2 [36-50]   43.2 [36-50] n=192
ssn                 153     68      1      98.6 [92-100]       44.4 [37-52]   90.5 [82-95] n=74
swift_bic           227     80     39       67.2 [58-75]       35.2 [29-42]   35.2 [29-42] n=227
tax_id                0      0     26         0.0 [0-13]                  -              - n=0
vehicle_reg           0      0      7         0.0 [0-35]                  -              - n=0
vin                   0      0     12         0.0 [0-24]                  -              - n=0
unmapped: date 8469 (0.4%), company 6342 (0.5%), time 1746 (0.1%), customer_id 195 (0.0%), employee_id 175 (0.0%), date_time 89 (0.0%), user_name 71 (9.9%)

ylemis (10000 docs, floor medium)
category           gold     tp     fp          precision             recall         recall valid
aadhaar            1757   1757      0    100.0 [100-100]    100.0 [100-100] 100.0 [100-100] n=1757
address            3649   1553      0    100.0 [100-100]       42.6 [41-44]   42.6 [41-44] n=3649
bank_account          0      0    500          0.0 [0-1]                  -              - n=0
email              1080   1080      0    100.0 [100-100]    100.0 [100-100] 100.0 [100-100] n=1080
gstin               405    405      0     100.0 [99-100]     100.0 [99-100] 100.0 [99-100] n=405
health_id           676    270      0     100.0 [99-100]       39.9 [36-44]   39.9 [36-44] n=676
ifsc                946    946      0    100.0 [100-100]    100.0 [100-100] 100.0 [100-100] n=946
pan                2567   2567      0    100.0 [100-100]    100.0 [100-100] 100.0 [100-100] n=2567
passport            405    405      0     100.0 [99-100]     100.0 [99-100] 100.0 [99-100] n=405
person_name        8244      0      0                  -          0.0 [0-0]      0.0 [0-0] n=8244
phone              3650   3650      0    100.0 [100-100]    100.0 [100-100] 100.0 [100-100] n=3650
upi_id             1081   1081      0    100.0 [100-100]    100.0 [100-100] 100.0 [100-100] n=1081
vehicle_reg         540    540      0     100.0 [99-100]     100.0 [99-100] 100.0 [99-100] n=540
voter_id            541    270      0     100.0 [99-100]       49.9 [46-54]   49.9 [46-54] n=541
negative documents with any finding: 0/1756 (0.0%) {'benign': '0/810', 'hard_negative': '0/946'}
unmapped: account_id 2433 (20.6%), location 2433 (0.0%), organization 1623 (0.0%)

maskara (17600 docs, floor medium)
category           gold     tp     fp          precision             recall         recall valid
aadhaar            5079    381      0     100.0 [99-100]          7.5 [7-8]   86.1 [83-89] n=440
address            2820   1802    202       89.9 [88-91]       63.9 [62-66]   63.9 [62-66] n=2820
card                698     20      0     100.0 [84-100]          2.9 [2-4] 100.0 [84-100] n=20
date_of_birth      2690   2436      0    100.0 [100-100]       90.6 [89-92]   90.6 [89-92] n=2690
driving_licence    1613   1553      0    100.0 [100-100]       96.3 [95-97]   96.3 [95-97] n=1613
email              3704   3664     40       98.9 [98-99]       98.9 [98-99] 100.0 [100-100] n=3663
health                0      0   1100          0.0 [0-0]                  -              - n=0
ip_address         1258    900      0    100.0 [100-100]       71.5 [69-74]   75.1 [73-78] n=1198
pan                1213    358      0     100.0 [99-100]       29.5 [27-32]   76.3 [72-80] n=469
passport            891    855      0    100.0 [100-100]       96.0 [94-97]   96.0 [94-97] n=891
person_name       10985   2300      0    100.0 [100-100]       20.9 [20-22]   20.9 [20-22] n=10985
phone              8279   4587     64       98.6 [98-99]       55.4 [54-56]   97.3 [97-98] n=3503
postal_code           0      0     15        0.0 [-0-20]                  -              - n=0
secret             2234   1896     29       98.5 [98-99]       84.9 [83-86]   84.9 [83-86] n=2234
ssn                 373    324      0     100.0 [99-100]       86.9 [83-90] 100.0 [99-100] n=324
upi_id             1483   1354      0    100.0 [100-100]       91.3 [90-93]   91.3 [90-93] n=1483
vehicle_reg        1585   1585      0    100.0 [100-100]    100.0 [100-100] 100.0 [100-100] n=1585
negative documents with any finding: 0/200 (0.0%) {'real_world_curated': '0/200'}
unmapped: USERNAME 1244 (0.0%)
