SentenceTransformer model

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+    "word_embedding_dimension": 768,
+    "pooling_mode_cls_token": false,
+    "pooling_mode_mean_tokens": true,
+    "pooling_mode_max_tokens": false,
+    "pooling_mode_mean_sqrt_len_tokens": false,
+    "pooling_mode_weightedmean_tokens": false,
+    "pooling_mode_lasttoken": false,
+    "include_prompt": true
+}

README.md

@@ -1,12 +1,167 @@
 ---
+library_name: sentence-transformers
+base_model:
+- DTAI-KULeuven/robbert-2023-dutch-base
+tags:
+- generated_from_trainer
+- transformers
 license: mit
 datasets:
 - clips/beir-nl-mmarco
 - clips/beir-nl-hotpotqa
 - clips/beir-nl-fever
-language:
-- nl
-base_model:
-- DTAI-KULeuven/robbert-2023-dutch-base
+language: nl
 pipeline_tag: sentence-similarity
----
+model-index:
+- name: RobBERT-2023-base-ft
+  results: []
+---
+
+# RobBERT-2023-base-ft
+
+RobBERT-2023-base-ft is a fine-tuned version of [DTAI-KULeuven/robbert-2023-dutch-base]( DTAI-KULeuven/robbert-2023-dutch-base).It demonstrates strong performance on MTEB-NL. If you’re looking for a state-of-the-art model of comparable size, you may also want to consider [clips/e5-base-trm-nl](https://huggingface.co/clips/e5-base-trm-nl)
+
+## Usage
+
+Below is an example to encode queries and passages from the MS-MARCO passage ranking dataset.
+
+```python
+import torch.nn.functional as F
+
+from torch import Tensor
+from transformers import AutoTokenizer, AutoModel
+
+
+def average_pool(last_hidden_states: Tensor,
+                 attention_mask: Tensor) -> Tensor:
+    last_hidden = last_hidden_states.masked_fill(~attention_mask[..., None].bool(), 0.0)
+    return last_hidden.sum(dim=1) / attention_mask.sum(dim=1)[..., None]
+
+
+# Each input text should start with "query: " or "passage: ".
+# For tasks other than retrieval, you can simply use the "query: " prefix.
+input_texts = [
+    'query: hoeveel eiwitten moet een vrouw eten',
+    'query: top definieer',
+    "passage: Als algemene richtlijn geldt dat de gemiddelde eiwitbehoefte voor vrouwen van 19 tot 70 jaar volgens de CDC 46 gram per dag bedraagt. Maar, zoals je in deze tabel kunt zien, moet je dit verhogen als je zwanger bent of traint voor een marathon. Bekijk de onderstaande tabel om te zien hoeveel eiwitten je dagelijks zou moeten eten.",
+    "passage: Definitie van top voor leerlingen Engels. : 1 het hoogste punt van een berg : de top van een berg. : 2 het hoogste niveau. : 3 een bijeenkomst of reeks bijeenkomsten tussen de leiders van twee of meer regeringen."
+]
+
+tokenizer = AutoTokenizer.from_pretrained('clips/robbert-2023-base-ft')
+model = AutoModel.from_pretrained('clips/robbert-2023-base-ft')
+
+# Tokenize the input texts
+batch_dict = tokenizer(input_texts, max_length=512, padding=True, truncation=True, return_tensors='pt')
+
+outputs = model(**batch_dict)
+embeddings = average_pool(outputs.last_hidden_state, batch_dict['attention_mask'])
+
+# normalize embeddings
+embeddings = F.normalize(embeddings, p=2, dim=1)
+scores = (embeddings[:2] @ embeddings[2:].T) * 100
+print(scores.tolist())
+```
+
+Below is an example for usage with sentence_transformers.
+```python
+from sentence_transformers import SentenceTransformer
+
+# Load the model from Hugging Face
+model = SentenceTransformer("clips/robbert-2023-base-ft")
+
+# Perform inference using encode_query/encode_document for retrieval,
+# or encode_query for general purpose embeddings. Prompt prefixes
+# are automatically added with these two methods.
+queries = [
+    'hoeveel eiwitten moet een vrouw eten',
+    'top definieer',
+]
+documents = [
+    'Als algemene richtlijn geldt dat de gemiddelde eiwitbehoefte voor vrouwen van 19 tot 70 jaar volgens de CDC 46 gram per dag bedraagt. Maar, zoals je in deze tabel kunt zien, moet je dit verhogen als je zwanger bent of traint voor een marathon. Bekijk de onderstaande tabel om te zien hoeveel eiwitten je dagelijks zou moeten eten.',
+    'Definitie van top voor leerlingen Engels. : 1 het hoogste punt van een berg : de top van een berg. : 2 het hoogste niveau. : 3 een bijeenkomst of reeks bijeenkomsten tussen de leiders van twee of meer regeringen.',
+]
+query_embeddings = model.encode_query(queries)
+document_embeddings = model.encode_document(documents)
+print(query_embeddings.shape, document_embeddings.shape)
+# (2, 768) (2, 768)
+
+similarities = model.similarity(query_embeddings, document_embeddings)
+print(similarities)
+# tensor([[0.7730, 0.2940],
+#         [0.3368, 0.7256]])
+```
+## Benchmark Evaluation
+Results on MTEB-NL (models introduced in [our paper](https://arxiv.org/abs/2509.12340) and the best model per size category are highlighted in bold):
+
+| Model                                 | Prm  | Cls      | MLCls    | PCls     | Rrnk     | Rtr      | Clust    | STS      | AvgD     | AvgT     |
+|---------------------------------------|------|----------|----------|----------|----------|----------|----------|----------|----------|----------|
+| **Num. Datasets (→)**                 |      | 12       | 3        | 2        | 1        | 12       | 8        | 2        | 40       |          |
+| **Supervised (small, <100M)**         |      |          |          |          |          |          |          |          |          |          |
+| **e5-small-v2-t2t**                   | 33M  | 53.7     | 38.5     | 74.5     | 85.9     | 45.0     | 24.1     | 74.3     | 46.9     | 56.6     |
+| **e5-small-v2-t2t-nl**                | 33M  | 55.3     | 40.9     | 74.9     | 86.0     | 49.9     | 28.0     | 74.1     | 49.8     | 58.4     |
+| **e5-small-trm**                      | 41M  | 56.3     | 43.5     | **76.5** | **87.3** | 53.1     | 28.2     | 74.2     | 51.4     | 59.9     |
+| **e5-small-trm-nl**                   | 41M  | **58.2** | **44.7** | 76.0     | 87.1     | **56.0** | **32.2** | **74.6** | **53.8** | **61.3** |
+| **Supervised (base, <305M)**          |      |          |          |          |          |          |          |          |          |          |
+| granite-embedding-107m-multilingual   | 107M | 53.9     | 41.8     | 70.1     | 84.7     | 50.2     | 29.8     | 68.4     | 49.4     | 57.0     |
+| **e5-base-v2-t2t**                    | 109M | 54.4     | 40.3     | 73.3     | 85.6     | 46.2     | 25.5     | 73.2     | 47.8     | 56.9     |
+| **e5-base-v2-t2t-nl**                 | 109M | 53.9     | 41.5     | 72.5     | 84.0     | 46.4     | 26.9     | 69.3     | 47.8     | 56.3     |
+| multilingual-e5-small                 | 118M | 56.3     | 43.5     | 76.5     | 87.1     | 53.1     | 28.2     | 74.2     | 51.4     | 59.8     |
+| paraphrase-multilingual-MiniLM-L12-v2 | 118M | 55.0     | 38.1     | 78.2     | 80.6     | 37.7     | 29.6     | 76.3     | 46.3     | 56.5     |
+| **RobBERT-2023-base-ft**              | 124M | 58.1     | 44.6     | 72.7     | 84.7     | 51.6     | 32.9     | 68.5     | 52.0     | 59.0     |
+| **e5-base-trm**                       | 124M | 58.1     | 44.4     | 76.7     | 88.3     | 55.8     | 28.1     | 74.9     | 52.9     | 60.9     |
+| **e5-base-trm-nl**                    | 124M | **59.6** | **45.9** | 78.4     | 87.5     | 56.5     | **34.3** | 75.8     | **55.0** | **62.6** |
+| potion-multilingual-128M              | 128M | 51.8     | 40.0     | 60.4     | 80.3     | 35.7     | 26.1     | 62.0     | 42.6     | 50.9     |
+| multilingual-e5-base                  | 278M | 58.2     | 44.4     | 76.7     | **88.4** | 55.8     | 27.7     | 74.9     | 52.8     | 60.9     |
+| granite-embedding-278m-multilingual   | 278M | 54.6     | 41.8     | 71.0     | 85.6     | 52.4     | 30.3     | 68.9     | 50.5     | 58.0     |
+| paraphrase-multilingual-mpnet-base-v2 | 278M | 58.1     | 40.5     | **81.9** | 82.3     | 41.4     | 30.8     | 79.3     | 49.2     | 59.2     |
+| Arctic-embed-m-v2.0                   | 305M | 54.4     | 42.6     | 66.6     | 86.2     | 51.8     | 26.5     | 64.9     | 49.1     | 56.1     |
+| gte-multilingual-base                 | 305M | 59.1     | 37.7     | 77.8     | 82.3     | **56.8** | 31.3     | **78.6** | 53.8     | 60.5     |
+| **Supervised (large, >305M)**         |      |          |          |          |          |          |          |          |          |          |
+| **e5-large-v2-t2t**                   | 335M | 55.7     | 41.4     | 75.7     | 86.6     | 49.9     | 25.5     | 74.0     | 49.5     | 58.4     |
+| **e5-large-v2-t2t-nl**                | 335M | 57.3     | 42.4     | 76.9     | 86.9     | 50.8     | 27.7     | 74.1     | 51.7     | 59.4     |
+| **RobBERT-2023-large-ft**             | 355M | 59.3     | 45.2     | 68.7     | 82.3     | 48.3     | 31.6     | 70.6     | 51.0     | 58.0     |
+| **e5-large-trm**                      | 355M | 60.2     | 45.4     | 80.3     | 90.3     | 59.0     | 28.7     | 78.8     | 55.1     | 63.3     |
+| **e5-large-trm-nl**                   | 355M | **62.2** | **48.0** | **81.4** | 87.2     | 58.2     | 35.6     | 78.2     | **57.0** | **64.4** |
+| multilingual-e5-large                 | 560M | 60.2     | 45.4     | 80.3     | **90.3** | 59.1     | 29.5     | 78.8     | 55.3     | 63.4     |
+| Arctic-embed-l-v2.0                   | 568M | 59.3     | 45.2     | 74.2     | 88.2     | 59.0     | 29.8     | 71.7     | 54.3     | 61.1     |
+| bge-m3                                | 568M | 60.7     | 44.2     | 78.3     | 88.7     | **60.0** | 29.2     | 78.1     | 55.4     | 63.1     |
+| jina-embeddings-v3                    | 572M | 61.7     | 38.9     | 76.8     | 78.5     | 59.1     | **38.9** | **84.8** | **57.0** | 62.7     |
+
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 1
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: constant_with_warmup
+- lr_scheduler_warmup_ratio: 0.25
+- num_epochs: 1.0
+
+### Framework versions
+
+- Transformers 4.56.1
+- Pytorch 2.7.1+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.0
+
+## Citation Information
+
+If you find our paper, benchmark or models helpful, please consider cite as follows:
+```latex
+@misc{banar2025mtebnle5nlembeddingbenchmark,
+      title={MTEB-NL and E5-NL: Embedding Benchmark and Models for Dutch}, 
+      author={Nikolay Banar and Ehsan Lotfi and Jens Van Nooten and Cristina Arhiliuc and Marija Kliocaite and Walter Daelemans},
+      year={2025},
+      eprint={2509.12340},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2509.12340}, 
+}
+```
+[//]: # (https://arxiv.org/abs/2509.12340)

config_sentence_transformers.json

@@ -0,0 +1,14 @@
+{
+  "model_type": "SentenceTransformer",
+  "__version__": {
+    "sentence_transformers": "5.1.0",
+    "transformers": "4.56.1",
+    "pytorch": "2.7.1+cu126"
+  },
+  "prompts": {
+    "query": "query: ",
+    "document": "passage: "
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+    "max_seq_length": 512,
+    "do_lower_case": false
+}