Update model card for Bengali

README.md

@@ -1,26 +1,92 @@
----
-pipeline_tag: zero-shot-image-classification
-language: ben
-tags:
-  - trimmed
-  - vision-language
-  - clip
-library_name: transformers
-base_model: facebook/metaclip-2-worldwide-m16-384
-base_model_relation: quantized
-datasets:
-  - Lumberjackk/fineweb-2-trimming
----
-
-# metaclip-2-worldwide-m16-384-ben-32768
-
-This model is a 82.6% smaller version of [facebook/metaclip-2-worldwide-m16-384](https://huggingface.co/facebook/metaclip-2-worldwide-m16-384)
-optimized for Bengali via text vocabulary trimming.
-
-## Model Statistics
-- **Original vocab size:** 901,629
-- **Trimmed vocab size:** 32,768
-- **Vocabulary reduction:** 96.4%
-- **Original params:** 538,547,201
-- **Trimmed params:** 93,690,369
-- **Size reduction:** 82.6%
+---
+pipeline_tag: sentence-similarity
+language: ben
+license: cc-by-nc-4.0
+tags:
+  - trimmed
+library_name: sentence-transformers
+base_model: facebook/metaclip-2-worldwide-m16-384
+base_model_relation: quantized
+datasets:
+  - lbourdois/fineweb-2-trimming
+---
+
+# metaclip-2-worldwide-m16-384-ben-32768
+This model is a **82.60% smaller** version of [facebook/metaclip-2-worldwide-m16-384](https://huggingface.co/facebook/metaclip-2-worldwide-m16-384) optimized for **Bengali** language via vocabulary size reduction using the [trimming](https://huggingface.co/blog/lbourdois/introduction-to-trimming) method.  
+This trimmed model should perform similarly to the original model with only 32,768 tokens and a much smaller memory footprint. However, it may not perform well for other languages as tokens not commonly used in the selected languages were removed from the vocabulary.
+
+## Model Statistics
+| Metric | Original | Trimmed | Reduction |
+|--------|----------|---------|-----------|
+| **Vocabulary size** | 901,629 tokens | 32,768 tokens | **96.37%** |
+| **Model size** | 538,547,201 params | 93,690,369 params | **82.60%** |
+
+![image](https://raw.githubusercontent.com/lbourdois/blog/refs/heads/master/assets/images/Trimming/metaclip-2-worldwide-m16-384-32768.png)
+
+## Mining Dataset Statistics
+- **Number of texts used for mining**: 200,000 texts  
+- **Dataset**: [lbourdois/fineweb-2-trimming](https://huggingface.co/datasets/lbourdois/fineweb-2-trimming)
+
+## Usage
+
+#### Transformers (zero-shot image classification)
+```python
+from transformers import pipeline
+
+# load pipeline
+image_classifier = pipeline(model="alphaedge-ai/metaclip-2-worldwide-m16-384-ben-32768", task="zero-shot-image-classification")
+
+# load image and candidate labels
+image = "http://images.cocodataset.org/val2017/000000039769.jpg"
+candidate_labels = ["Potential label 1 in Bengali", "Potential label 2 in Bengali", "Potential label 3 in Bengali", "Potential label 4 in Bengali"]
+
+# run inference
+outputs = image_classifier(image, candidate_labels)
+print(outputs)
+```
+
+#### Sentence-transformers (texts-images similarity)
+```python
+from sentence_transformers import SentenceTransformer
+
+model = SentenceTransformer("alphaedge-ai/metaclip-2-worldwide-m16-384-ben-32768")
+
+images = [
+    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/car.jpg",
+    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg",
+    "https://huggingface.co/datasets/huggingface/cats-image/resolve/main/cats_image.jpeg"
+]
+texts = ["Text 1 in Bengali", "Text 2 in Bengali", "Text 3 in Bengali", "Text 4 in Bengali"]
+
+image_embeddings = model.encode(images)
+text_embeddings = model.encode(texts)
+print(image_embeddings.shape, text_embeddings.shape)
+
+similarities = model.similarity(image_embeddings, text_embeddings)
+print(similarities)
+```
+
+## Citations
+
+#### Meta CLIP 2
+```
+@misc{chuang2025metaclip2worldwide,
+      title={Meta CLIP 2: A Worldwide Scaling Recipe}, 
+      author={Yung-Sung Chuang and Yang Li and Dong Wang and Ching-Feng Yeh and Kehan Lyu and Ramya Raghavendra and James Glass and Lifei Huang and Jason Weston and Luke Zettlemoyer and Xinlei Chen and Zhuang Liu and Saining Xie and Wen-tau Yih and Shang-Wen Li and Hu Xu},
+      year={2025},
+      eprint={2507.22062},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2507.22062}, 
+}
+```
+
+#### Trimming blog post
+```
+@misc{hf_blogpost_trimming,
+      title={Introduction to Trimming}, 
+      author={Loïck BOURDOIS and Tom AARSEN and Bram VANROY and Christopher AKIKI and Woojun JUNG and Manuel ROMERO and Prithiv SAKTHI},
+      year={2026},
+      url={https://huggingface.co/blog/lbourdois/introduction-to-trimming}, 
+}
+```