SecurePII-Bench
Collection
262 items • Updated
config string | train_flags string | base_model string | seeds list | eval_flags string | batch_size int64 | patience int64 | datasets dict | runs dict | summary_table list | latex_row string | generated_at timestamp[s] | note string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
cw | --loss-class-weight | fastino/gliner2-multi-v1 | [
42,
43,
44
] | --threshold 0.7 --use-desc --schema-case original --no-extra-val | 16 | 2 | {
"fin": {
"epochs": 8,
"max_val": null,
"repo": "quynong/fin",
"eval_split": "test",
"n_labels": 4
},
"mit_restaurant": {
"epochs": 6,
"max_val": null,
"repo": "quynong/mit_restaurant",
"eval_split": "test",
"n_labels": 8
},
"conll2003": {
"epochs": 6,
"max_val... | {
"fin-cw-seed42": {
"status": "ok",
"dataset": "fin",
"config": "cw",
"seed": 42,
"repo": "AITeamUIT/gliner2-ner-fin-cw-seed42",
"train_flags": "--loss-class-weight",
"eval_flags": "--threshold 0.7 --use-desc --schema-case original --no-extra-val",
"epochs": 8,
"batch_size": 16,
... | [
{
"mode": "lenient",
"dataset": "conll2003",
"n_seeds": 3,
"macro_mean": 88.07,
"macro_std": 0.98,
"micro_mean": 89.31,
"micro_std": 0.95,
"macro_f1": "88.07 +/- 0.98",
"micro_f1": "89.31 +/- 0.95"
},
{
"mode": "lenient",
"dataset": "ontonotes5",
"n_seeds": 3,
... | Class-weighted & $ 86.06_{\pm 0.95} $ & $ 77.47_{\pm 0.51} $ & $ 38.15_{\pm 2.40} $ & $ 81.55_{\pm 0.26} $ & 70.81 \\ | 2026-08-22T14:20:16 | Chay lai class-weighted 3 seed de dong nay dung cung giao thuc eval voi 5 config loss ablation (--use-desc --schema-case original, threshold 0.7). Run cu AITeamUIT/gliner2-ner-*-cw-20260706 la 1 seed, eval non-desc. |
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Class-weighted loss — 3 seed x 4 dataset
- train:
--loss-class-weighttrenfastino/gliner2-multi-v1, batch 16, patience 2, seeds [42, 43, 44] - eval :
--threshold 0.7 --use-desc --schema-case original --no-extra-val(giong loss ablation)
| dataset | epochs | eval split | strict macro F1 | lenient micro F1 |
|---|---|---|---|---|
| conll2003 | 6 | test | 86.06 +/- 0.95 | 89.31 +/- 0.95 |
| ontonotes5 | 3 | test | 77.47 +/- 0.51 | 90.30 +/- 0.06 |
| fin | 8 | test | 38.15 +/- 2.40 | 75.92 +/- 1.36 |
| mit_restaurant | 6 | test | 81.55 +/- 0.26 | 88.25 +/- 0.66 |
Dong cho tab:macro-f1 (strict macro F1)
Class-weighted & $ 86.06_{\pm 0.95} $ & $ 77.47_{\pm 0.51} $ & $ 38.15_{\pm 2.40} $ & $ 81.55_{\pm 0.26} $ & 70.81 \\
Checkpoint tung run
fin-cw-seed42-> AITeamUIT/gliner2-ner-fin-cw-seed42 (strict macro F1 38.21)fin-cw-seed43-> AITeamUIT/gliner2-ner-fin-cw-seed43 (strict macro F1 35.72)fin-cw-seed44-> AITeamUIT/gliner2-ner-fin-cw-seed44 (strict macro F1 40.51)mit_restaurant-cw-seed42-> AITeamUIT/gliner2-ner-mit_restaurant-cw-seed42 (strict macro F1 81.45)mit_restaurant-cw-seed43-> AITeamUIT/gliner2-ner-mit_restaurant-cw-seed43 (strict macro F1 81.35)mit_restaurant-cw-seed44-> AITeamUIT/gliner2-ner-mit_restaurant-cw-seed44 (strict macro F1 81.84)conll2003-cw-seed42-> AITeamUIT/gliner2-ner-conll2003-cw-seed42 (strict macro F1 85.09)conll2003-cw-seed43-> AITeamUIT/gliner2-ner-conll2003-cw-seed43 (strict macro F1 86.12)conll2003-cw-seed44-> AITeamUIT/gliner2-ner-conll2003-cw-seed44 (strict macro F1 86.98)ontonotes5-cw-seed42-> AITeamUIT/gliner2-ner-ontonotes5-cw-seed42 (strict macro F1 77.61)ontonotes5-cw-seed43-> AITeamUIT/gliner2-ner-ontonotes5-cw-seed43 (strict macro F1 77.89)ontonotes5-cw-seed44-> AITeamUIT/gliner2-ner-ontonotes5-cw-seed44 (strict macro F1 76.90)
generated 2026-08-22T14:20:16
- Downloads last month
- 15