Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+banner.png filter=lfs diff=lfs merge=lfs -text

0_Transformer/config.json

@@ -0,0 +1,23 @@
+{
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertModel"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "hidden_dim": 3072,
+  "initializer_range": 0.02,
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "transformers_version": "4.57.6",
+  "vocab_size": 30522
+}

0_Transformer/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94a6e24feffa39d1bb68301894d7df0d8f79303246e2b2f41ce2442145164d7d
+size 265462608

0_Transformer/sentence_bert_config.json

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

0_Transformer/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

0_Transformer/tokenizer_config.json

@@ -0,0 +1,56 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "DistilBertTokenizer",
+  "unk_token": "[UNK]"
+}

1_Pooling/config.json

@@ -0,0 +1,9 @@
+{
+  "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
+}

2_Dense/config.json

@@ -0,0 +1,6 @@
+{
+  "in_features": 768,
+  "out_features": 768,
+  "bias": true,
+  "activation_function": "torch.nn.modules.activation.GELU"
+}

2_Dense/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:878c5b354f0f87e2f3bbde25c161790eb0332d496b34dc7dce503adb307e9645
+size 2364309

3_Dense/config.json

@@ -0,0 +1,6 @@
+{
+  "in_features": 768,
+  "out_features": 2048,
+  "bias": true,
+  "activation_function": "torch.nn.modules.linear.Identity"
+}

3_Dense/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ae433824a6c8d050d21e70f11118f3269d765a87790d19343bd5fa65334083a
+size 6301589

README.md

@@ -0,0 +1,114 @@
+---
+library_name: sentence-transformers
+pipeline_tag: sentence-similarity
+tags:
+- sentence-transformers
+- feature-extraction
+- sentence-similarity
+- visual-document-retrieval
+- cross-modal-distillation
+- knowledge-distillation
+- nanovdr
+base_model: distilbert/distilbert-base-uncased
+language:
+- en
+license: apache-2.0
+datasets:
+- openbmb/VisRAG-Ret-Train-Synthetic-data
+- openbmb/VisRAG-Ret-Train-In-domain-data
+- vidore/colpali_train_set
+- llamaindex/vdr-multilingual-train
+model-index:
+- name: NanoVDR-S
+  results:
+  - task:
+      type: retrieval
+    dataset:
+      name: ViDoRe v1
+      type: vidore/vidore-benchmark-667173f98e70a1c0fa4d
+    metrics:
+    - name: NDCG@5
+      type: ndcg_at_5
+      value: 82.2
+  - task:
+      type: retrieval
+    dataset:
+      name: ViDoRe v2
+      type: vidore/vidore-benchmark-v2
+    metrics:
+    - name: NDCG@5
+      type: ndcg_at_5
+      value: 60.5
+---
+
+<p align="center">
+  <img width="560" src="banner.png" alt="NanoVDR"/>
+</p>
+
+# NanoVDR-S
+
+**English-only baseline variant.** For production use (especially with multilingual queries), we recommend **[NanoVDR-S-Multi](https://huggingface.co/nanovdr/NanoVDR-S-Multi)**.
+
+NanoVDR-S is a 69M-parameter text-only query encoder for visual document retrieval, trained via asymmetric cross-modal distillation from [Qwen3-VL-Embedding-2B](https://huggingface.co/Qwen/Qwen3-VL-Embedding-2B). It uses DistilBERT + a 2-layer MLP projector to encode text queries into the teacher's embedding space.
+
+## Results
+
+| Model | Params | ViDoRe v1 | ViDoRe v2 | ViDoRe v3 | Avg Retention |
+|-------|--------|-----------|-----------|-----------|---------------|
+| Qwen3-VL-Emb (Teacher) | 2.0B | 84.3 | 65.3 | 50.0 | — |
+| **NanoVDR-S** | **69M** | **82.2** | **60.5** | **43.5** | **92.4%** |
+| NanoVDR-S-Multi | 69M | 82.2 | 61.9 | 46.5 | 95.1% |
+
+<sub>NDCG@5 (×100). Retention = Student / Teacher averaged across v1/v2/v3.</sub>
+
+## Usage
+
+> **Prerequisite:** Documents must be indexed offline using [Qwen3-VL-Embedding-2B](https://huggingface.co/Qwen/Qwen3-VL-Embedding-2B) (the teacher model). See the [NanoVDR-S-Multi model page](https://huggingface.co/nanovdr/NanoVDR-S-Multi#prerequisites-document-indexing-with-teacher-model) for a complete indexing guide.
+
+```python
+from sentence_transformers import SentenceTransformer
+import numpy as np
+
+# doc_embeddings: (N, 2048) from teacher indexing (see prerequisite above)
+
+model = SentenceTransformer("nanovdr/NanoVDR-S")
+query_embeddings = model.encode(["What was the revenue growth in Q3?"])  # (1, 2048)
+
+scores = query_embeddings @ doc_embeddings.T
+top_k_indices = np.argsort(scores[0])[-5:][::-1]
+```
+
+## Training Details
+
+| | Value |
+|--|-------|
+| Architecture | DistilBERT (66M) + MLP projector (768 → 768 → 2048, 2.4M) = 69M |
+| Objective | Pointwise cosine alignment with teacher query embeddings |
+| Data | 711K query-document pairs |
+| Epochs / lr | 20 / 2e-4 |
+| Training cost | ~10 GPU-hours (1× H200) |
+| CPU query latency | 51 ms |
+
+## All NanoVDR Models
+
+| Model | Backbone | Params | v1 | v2 | v3 | Retention |
+|-------|----------|--------|----|----|----| ----------|
+| **[NanoVDR-S-Multi](https://huggingface.co/nanovdr/NanoVDR-S-Multi)** | **DistilBERT** | **69M** | **82.2** | **61.9** | **46.5** | **95.1%** |
+| [NanoVDR-S](https://huggingface.co/nanovdr/NanoVDR-S) | DistilBERT | 69M | 82.2 | 60.5 | 43.5 | 92.4% |
+| [NanoVDR-M](https://huggingface.co/nanovdr/NanoVDR-M) | BERT-base | 112M | 82.1 | 62.2 | 44.7 | 94.0% |
+| [NanoVDR-L](https://huggingface.co/nanovdr/NanoVDR-L) | ModernBERT | 151M | 82.4 | 61.5 | 44.2 | 93.4% |
+
+## Citation
+
+```bibtex
+@article{nanovdr2026,
+  title={NanoVDR: Distilling a 2B Vision-Language Retriever into a 70M Text-Only Encoder for Visual Document Retrieval},
+  author={Liu, Zhuchenyang and Zhang, Yao and Xiao, Yu},
+  journal={arXiv preprint arXiv:2502.XXXXX},
+  year={2026}
+}
+```
+
+## License
+
+Apache 2.0

config.json

@@ -0,0 +1,23 @@
+{
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertModel"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "hidden_dim": 3072,
+  "initializer_range": 0.02,
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "transformers_version": "4.57.6",
+  "vocab_size": 30522
+}

config_sentence_transformers.json

@@ -0,0 +1,5 @@
+{
+  "prompts": {},
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

modules.json

@@ -0,0 +1,32 @@
+[
+  {
+    "idx": 0,
+    "name": "0_Transformer",
+    "path": "0_Transformer",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1_Pooling",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2_Dense",
+    "path": "2_Dense",
+    "type": "sentence_transformers.models.Dense"
+  },
+  {
+    "idx": 3,
+    "name": "3_Dense",
+    "path": "3_Dense",
+    "type": "sentence_transformers.models.Dense"
+  },
+  {
+    "idx": 4,
+    "name": "4_Normalize",
+    "path": "4_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,56 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "DistilBertTokenizer",
+  "unk_token": "[UNK]"
+}