README.md

---
license: apache-2.0
base_model: answerdotai/ModernBERT-base
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- biomedical
- embeddings
- life-sciences
- scientific-text
- SODA-VEC
- EMBO
datasets:
- EMBO/soda-vec-data-full_pmc_title_abstract_paired
metrics:
- cosine-similarity
---

# Dot Only Model

## Model Description

SODA-VEC embedding model trained with dot product loss only. This model uses normalized embeddings with only contrastive learning (dot product) to learn biomedical text representations.

This model is part of the **SODA-VEC** (Scientific Open Domain Adaptation for Vector Embeddings) project, which focuses on creating high-quality embedding models for biomedical and life sciences text.

**Key Features:**
- Trained on **26.5M biomedical title-abstract pairs** from PubMed Central
- Based on **ModernBERT-base** architecture
- Optimized for **biomedical text similarity** and **semantic search**
- Produces **768-dimensional embeddings** with mean pooling

## Training Details

### Training Data

- **Dataset**: [`EMBO/soda-vec-data-full_pmc_title_abstract_paired`](https://huggingface.co/datasets/EMBO/soda-vec-data-full_pmc_title_abstract_paired)
- **Size**: 26,473,900 training pairs
- **Source**: Complete PubMed Central baseline (July 2024)
- **Format**: Paired title-abstract examples optimized for contrastive learning

### Training Procedure

**Loss Function**: Dot Only: normalized embeddings with only dot product loss (diagonal + off-diagonal)

**Coefficients**: dot=1.0
**Base Model**: `answerdotai/ModernBERT-base`

**Training Configuration:**
- **GPUs**: 4
- **Batch Size per GPU**: 16
- **Gradient Accumulation**: 4
- **Effective Batch Size**: 256
- **Learning Rate**: 2e-05
- **Warmup Steps**: 100
- **Pooling Strategy**: mean
- **Epochs**: 1 (full dataset pass)

**Training Command:**
```bash
python scripts/soda-vec-train.py --config dot_only --coeff_dot 1 --push_to_hub --hub_org EMBO --save_limit 5
```

### Model Architecture

- **Base Architecture**: ModernBERT-base (12 layers, 768 hidden size)
- **Pooling**: Mean pooling over token embeddings
- **Output Dimension**: 768
- **Normalization**: L2-normalized embeddings (for VICReg-based models)

## Usage

### Using Sentence-Transformers

```python
from sentence_transformers import SentenceTransformer

# Load the model
model = SentenceTransformer("EMBO/dot_only")

# Encode sentences
sentences = [
    "CRISPR-Cas9 gene editing in human cells",
    "Genome editing using CRISPR technology"
]

embeddings = model.encode(sentences)
print(f"Embedding shape: {embeddings.shape}")

# Compute similarity
from sentence_transformers.util import cos_sim
similarity = cos_sim(embeddings[0], embeddings[1])
print(f"Similarity: {similarity.item():.4f}")
```

### Using Hugging Face Transformers

```python
from transformers import AutoTokenizer, AutoModel
import torch
import torch.nn.functional as F

# Load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("EMBO/dot_only")
model = AutoModel.from_pretrained("EMBO/dot_only")

# Encode sentences
sentences = [
    "CRISPR-Cas9 gene editing in human cells",
    "Genome editing using CRISPR technology"
]

inputs = tokenizer(sentences, padding=True, truncation=True, return_tensors="pt")
with torch.no_grad():
    outputs = model(**inputs)
    
# Mean pooling
embeddings = outputs.last_hidden_state.mean(dim=1)

# Normalize (for VICReg models)
embeddings = F.normalize(embeddings, p=2, dim=1)

# Compute similarity
similarity = F.cosine_similarity(embeddings[0:1], embeddings[1:2])
print(f"Similarity: {similarity.item():.4f}")
```

<!-- ## Evaluation

The model has been evaluated on comprehensive biomedical benchmarks including:

- **Journal-Category Classification**: Matching journals to BioRxiv subject categories
- **Title-Abstract Similarity**: Discriminating between related and unrelated paper pairs
- **Field-Specific Separability**: Distinguishing between different biological fields
- **Semantic Search**: Retrieval quality on biomedical text corpora

For detailed evaluation results, see the [SODA-VEC benchmark notebooks](https://github.com/EMBO/soda-vec).
 -->
## Intended Use

This model is designed for:

- **Biomedical Semantic Search**: Finding relevant papers, abstracts, or text passages
- **Scientific Text Similarity**: Computing similarity between biomedical texts
<!-- - **Information Retrieval**: Building search systems for scientific literature
- **Downstream Tasks**: As a base for fine-tuning on specific biomedical tasks
- **Research Applications**: Academic and research use in life sciences
 -->
## Limitations

- **Domain Specificity**: Optimized for biomedical and life sciences text; may not perform as well on general domain text
- **Language**: English only
- **Text Length**: Optimized for titles and abstracts; longer documents may require chunking
- **Bias**: Inherits biases from the training data (PubMed Central corpus)

## Citation

If you use this model, please cite:

```bibtex
@software{soda_vec,
  title = {SODA-VEC: Scientific Open Domain Adaptation for Vector Embeddings},
  author = {EMBO},
  year = {2024},
  url = {https://github.com/source-data/soda-vec}
}
```

## Model Card Contact

For questions or issues, please open an issue on the [SODA-VEC GitHub repository](https://github.com/EMBO/soda-vec).

---

**Model Card Generated**: 2025-11-10