canpolatbulbul commited on
Commit
1b9a056
·
verified ·
1 Parent(s): 3e2c85d

Add deprecation notice — see lexglue-* models for updated results

Browse files
Files changed (1) hide show
  1. README.md +11 -0
README.md CHANGED
@@ -18,6 +18,17 @@ metrics:
18
  pipeline_tag: text-classification
19
  ---
20
 
 
 
 
 
 
 
 
 
 
 
 
21
  # deberta-unfair-tos-augmented
22
 
23
  Best performing model - DeBERTa trained with augmented data for UNFAIR-ToS classification
 
18
  pipeline_tag: text-classification
19
  ---
20
 
21
+ > ⚠️ **DEPRECATED** — This model used a non-standard training methodology (Focal Loss + class weighting) and was evaluated on a synthetic test set. For accurate, reproducible results on the official LexGLUE benchmark, use our updated models:
22
+ >
23
+ > | Model | μ-F1 | m-F1 |
24
+ > |-------|------|------|
25
+ > | **[lexglue-roberta-unfair-tos](https://huggingface.co/Agreemind/lexglue-roberta-unfair-tos)** | **96.1** | **84.4** |
26
+ > | [lexglue-legalbert-unfair-tos](https://huggingface.co/Agreemind/lexglue-legalbert-unfair-tos) | 96.0 | 84.1 |
27
+ > | [lexglue-deberta-unfair-tos](https://huggingface.co/Agreemind/lexglue-deberta-unfair-tos) | 95.6 | 82.2 |
28
+ > | [lexglue-legalbert-small-unfair-tos](https://huggingface.co/Agreemind/lexglue-legalbert-small-unfair-tos) | 95.0 | 78.5 |
29
+
30
+
31
+
32
  # deberta-unfair-tos-augmented
33
 
34
  Best performing model - DeBERTa trained with augmented data for UNFAIR-ToS classification