---
language:
- de
- fr
- it
license: cc0-1.0
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- swiss-law
- legal-retrieval
- dense-retrieval
base_model: harrier-meta-distilled-v2
datasets:
- voilaj/swiss-caselaw
---

# harrier-semantic-v1

**Semantic retrieval model for Swiss court decisions** — fine-tuned on 55,000 (Sachverhalt → cited decision) pairs.

Understands natural language descriptions of legal situations in German, French, and Italian
and retrieves relevant Swiss Federal Court (BGer/BGE) decisions.

## Performance (Semantic Benchmark, 800 items)

| Metric | Baseline | harrier-semantic-v1 |
|--------|----------|---------------------|
| MRR@10 | 0.0823 | **0.1501** (+82%) |
| Recall@10 | — | 0.0744 |
| Hit@1 | — | 0.0925 |

## Usage

```python
from sentence_transformers import SentenceTransformer

model = SentenceTransformer("ArneH/harrier-semantic-v1")

# Describe a legal situation in natural language
query = "Arbeitnehmer wurde während Krankheit fristlos entlassen"
embedding = model.encode(query, normalize_embeddings=True)
```

## MCP Server (Claude Desktop / Claude Code)

Use with the Swiss Legal Semantic MCP server for offline semantic + keyword search
over 963,000+ Swiss court decisions:

```bash
pip install mcp sentence-transformers numpy huggingface_hub
python server.py --setup  # auto-downloads model + embeddings
```

Pre-computed corpus embeddings: [ArneH/swiss-caselaw-embeddings](https://huggingface.co/datasets/ArneH/swiss-caselaw-embeddings)

## Training

- **Base model**: harrier-meta-distilled-v2 (0.6B Qwen3-based)
- **Training pairs**: 55,000 (Sachverhalt → cited decisions, DE/FR/IT)
- **Loss**: MultipleNegativesRankingLoss
- **Epochs**: 3, Batch: 512 (2× B200 GPU, DDP)
- **Data source**: [voilaj/swiss-caselaw](https://huggingface.co/datasets/voilaj/swiss-caselaw) (963k decisions)