Update README.md
Browse files
README.md
CHANGED
|
@@ -39,6 +39,13 @@ k=X means that as a training data for this evaluation, we took only X examples f
|
|
| 39 |
| RoBERTa-base + NER-BERT pre-training | 32.3 | 50.9 | 61.9 | 67.6 |
|
| 40 |
| NuNER v1.0 | **39.4** | **59.6** | **67.8** | **71.5** |
|
| 41 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
Read more about evaluation protocol & datasets in our [paper](https://arxiv.org/abs/2402.15343) and [blog post](https://www.numind.ai/blog/a-foundation-model-for-entity-recognition).
|
| 43 |
|
| 44 |
## Usage
|
|
|
|
| 39 |
| RoBERTa-base + NER-BERT pre-training | 32.3 | 50.9 | 61.9 | 67.6 |
|
| 40 |
| NuNER v1.0 | **39.4** | **59.6** | **67.8** | **71.5** |
|
| 41 |
|
| 42 |
+
NuNER v1.0 has similar performance to 7B LLMs (70 times bigger that NuNER v1.0) created specifically for NER task.
|
| 43 |
+
|
| 44 |
+
| Model | k=8~16| k=64~128 |
|
| 45 |
+
|----------|----------|----------|
|
| 46 |
+
| UniversalNER (7B) | 57.89 ± 4.34 | 71.02 ± 1.53 |
|
| 47 |
+
| NuNER v1.0 (100M) | 58.75 ± 0.93 | 70.30 ± 0.35 |
|
| 48 |
+
|
| 49 |
Read more about evaluation protocol & datasets in our [paper](https://arxiv.org/abs/2402.15343) and [blog post](https://www.numind.ai/blog/a-foundation-model-for-entity-recognition).
|
| 50 |
|
| 51 |
## Usage
|