Update README.md
Browse files
README.md
CHANGED
|
@@ -44,7 +44,8 @@ k=X means that as training data for this evaluation, we took only X examples for
|
|
| 44 |
|----------|----------|----------|----------|----------|
|
| 45 |
| RoBERTa-base | 24.5 | 44.7 | 58.1 | 65.4
|
| 46 |
| RoBERTa-base + NER-BERT pre-training | 32.3 | 50.9 | 61.9 | 67.6 |
|
| 47 |
-
| NuNER
|
|
|
|
| 48 |
|
| 49 |
NuNER v1.0 has similar performance to 7B LLMs (70 times bigger than NuNER v1.0) created specifically for the NER task.
|
| 50 |
|
|
|
|
| 44 |
|----------|----------|----------|----------|----------|
|
| 45 |
| RoBERTa-base | 24.5 | 44.7 | 58.1 | 65.4
|
| 46 |
| RoBERTa-base + NER-BERT pre-training | 32.3 | 50.9 | 61.9 | 67.6 |
|
| 47 |
+
| NuNER v0.1 | 34.3 | 54.6 | 64.0 | 68.7 |
|
| 48 |
+
| NuNER v1.0 | 39.4 | 59.6 | 67.8 | 71.5 |
|
| 49 |
|
| 50 |
NuNER v1.0 has similar performance to 7B LLMs (70 times bigger than NuNER v1.0) created specifically for the NER task.
|
| 51 |
|