---
license: apache-2.0
language:
  - ca
  - es
  - en
base_model: BSC-LT/salamandra-7b-instruct-tools
library_name: transformers
pipeline_tag: text-generation
tags:
  - query-parsing
  - semantic-search
  - structured-output
  - json-generation
  - multilingual
  - catalan
  - spanish
  - LoRA
  - fine-tuned
  - AINA
  - R&D
datasets:
  - SIRIS-Lab/impuls-query-parsing
metrics:
  - accuracy
model-index:
  - name: IMPULS-Salamandra-7B-Query-Parser
    results:
      - task:
          type: text-generation
          name: Query Parsing
        metrics:
          - name: JSON Validity
            type: accuracy
            value: 1.0
          - name: Strict Accuracy
            type: accuracy
            value: 0.51
          - name: Relaxed Accuracy
            type: accuracy
            value: 0.65
          - name: Language Match
            type: accuracy
            value: 0.87
---

# IMPULS-Salamandra-7B-Query-Parser

A fine-tuned version of [BSC-LT/salamandra-7b-instruct-tools](https://huggingface.co/BSC-LT/salamandra-7b-instruct-tools) for converting natural language queries into structured JSON for R&D project semantic search.

## Model Description

This model was developed as part of the **IMPULS project** (AINA Challenge 2024), a collaboration between [SIRIS Academic](https://sirisacademic.com/) and [Generalitat de Catalunya](https://web.gencat.cat/) to build a multilingual semantic search system for Catalonia's R&D ecosystem (RIS3-MCAT platform).

The model converts natural language queries in **Catalan, Spanish, and English** into structured JSON containing:
- **Semantic query**: Core thematic content for vector search
- **Filters**: Structured metadata (funding programme, year range, location, organization type)
- **Query rewrite**: Human-readable interpretation of the query
- **Metadata**: Language detection and processing notes

### Example

**Input (Catalan):**
```
projectes d'IA en salut finançats per H2020 des de 2020
```

**Output:**
```json
{
  "doc_type": "projects",
  "filters": {
    "programme": "Horizon 2020",
    "year": ">=2020"
  },
  "organisations": [],
  "semantic_query": "intel·ligència artificial salut",
  "query_rewrite": "Projectes sobre IA en salut del programa H2020 des de 2020",
  "meta": {
    "lang": "CA"
  }
}
```

## Training Details

### Base Model
- **Model**: [BSC-LT/salamandra-7b-instruct-tools](https://huggingface.co/BSC-LT/salamandra-7b-instruct-tools)
- **Architecture**: LlamaForCausalLM (7B parameters)

### Fine-tuning Method
- **Technique**: LoRA (Low-Rank Adaptation)
- **Trainable parameters**: ~1% of total (~50MB adapter)

### LoRA Configuration
| Parameter | Value |
|-----------|-------|
| Rank (r) | 16 |
| Alpha | 32 |
| Dropout | 0.05 |
| Target modules | q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj |

### Training Hyperparameters
| Parameter | Value |
|-----------|-------|
| Epochs | 3 |
| Batch size | 16 (effective) |
| Learning rate | 2e-4 |
| Sequence length | 2048 |
| Precision | FP16 (mixed) |
| Optimizer | AdamW |
| LR scheduler | Cosine |
| Warmup ratio | 0.1 |

### Training Data

- **Dataset**: [SIRIS-Lab/impuls-query-parsing](https://huggingface.co/datasets/SIRIS-Lab/impuls-query-parsing)
- **Training split**: 682 multilingual queries (synthetic, template-generated)
- **Language distribution**: ~33% Catalan, ~33% Spanish, ~33% English
- **Query types**: Discover (88%), Quantify (12%)

### Evaluation Data

- **Test split**: 100 real queries from domain experts (SIRIS Academic)
- **Annotation**: Manual gold-standard JSON for each query

## Evaluation Results

### Overall Performance

| Metric | Base Model | Fine-tuned |
|--------|------------|------------|
| JSON Validity | 100% | **100%** |
| Strict Accuracy | 15% | **51%** |
| Relaxed Accuracy | 29% | **65%** |
| Language Match | 53% | **87%** |
| Semantic Query Accuracy | 44% | **86%** |

### Component-level Accuracy

| Component | Accuracy |
|-----------|----------|
| Programme (H2020, FEDER, etc.) | 96% |
| Year extraction | 98% |
| Location | 91% |
| Organizations | 77% |
| Semantic Query | 86% |

### Performance by Language

| Language | Relaxed Accuracy |
|----------|------------------|
| English | 72% |
| Catalan | 64% |
| Spanish | 52% |

### Comparison with Other Models

| Model | Strict Accuracy | Relaxed Accuracy | JSON Valid |
|-------|-----------------|------------------|------------|
| **Salamandra-7B (ours)** | **51%** | **65%** | 100% |
| Qwen 2.5-7B | 47% | 65% | 100% |
| Mistral-7B | 24% | 55% | 100% |

## Usage

### Basic Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "SIRIS-Lab/impuls-salamandra-7b-query-parser"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16,
    device_map="auto"
)

# System prompt (simplified version)
system_prompt = """Convert natural language queries into structured JSON for R&D project search.
Output only valid JSON with the required schema."""

query = "projectes d'hidrogen finançats per H2020 des de 2020"

messages = [
    {"role": "system", "content": system_prompt},
    {"role": "user", "content": query}
]

input_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(input_text, return_tensors="pt").to(model.device)

with torch.no_grad():
    outputs = model.generate(
        **inputs,
        max_new_tokens=512,
        temperature=0.1,
        do_sample=True
    )

response = tokenizer.decode(outputs[0][inputs['input_ids'].shape[1]:], skip_special_tokens=True)
print(response)
```

### With 4-bit Quantization (Recommended for limited VRAM)

```python
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
import torch

quantization_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_compute_dtype=torch.float16,
    bnb_4bit_quant_type="nf4"
)

model = AutoModelForCausalLM.from_pretrained(
    "SIRIS-Lab/impuls-salamandra-7b-query-parser",
    quantization_config=quantization_config,
    device_map="auto"
)
# Reduces memory from ~14GB to ~3.5GB
```

## Output Schema

```json
{
  "doc_type": "projects",
  "filters": {
    "programme": "string | null",
    "funding_level": "string | null", 
    "year": "string | null",
    "location": "string | null",
    "location_level": "region | province | country | null"
  },
  "organisations": [
    {
      "type": "university | research_center | hospital | company | null",
      "name": "string | null",
      "location": "string | null",
      "location_level": "string | null"
    }
  ],
  "semantic_query": "string | null",
  "query_rewrite": "string",
  "meta": {
    "lang": "CA | ES | EN",
    "notes": "string | null"
  }
}
```

## Hardware Requirements

| Configuration | VRAM Required |
|---------------|---------------|
| FP16 (full precision) | ~14 GB |
| 4-bit quantization | ~3.5 GB |

**Recommended**: GPU with 24GB+ VRAM (A100) or 4-bit quantization on consumer GPUs.

## Limitations

- **Domain-specific**: Optimized for R&D project search queries; may not generalize well to other domains
- **Schema-bound**: Outputs follow a fixed JSON schema; cannot handle arbitrary structured formats
- **Language coverage**: Best performance on Catalan and English; Spanish accuracy is lower
- **Complex queries**: Struggles with queries requiring numerical aggregation or ranking operations

## Intended Use

This model is designed for:
- R&D project discovery platforms (RIS3CAT, Horizon Europe portals)
- Scientific literature search systems
- Multilingual semantic search applications
- Query understanding in Catalan, Spanish, and English

## Ethical Considerations

- The model was trained on synthetic queries generated from templates and real queries from domain experts
- No personal or sensitive data was used in training
- The model is intended for search query parsing and does not generate harmful content

## Citation

If you use this model, please cite:

```bibtex
@misc{impuls-salamandra-2024,
  author = {SIRIS Academic},
  title = {IMPULS-Salamandra-7B-Query-Parser: Multilingual Query Parsing for R&D Semantic Search},
  year = {2024},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/SIRIS-Lab/impuls-salamandra-7b-query-parser}}
}
```

## Acknowledgments

- **[Barcelona Supercomputing Center (BSC)](https://www.bsc.es/)** - For the Salamandra base model and AINA infrastructure
- **[Generalitat de Catalunya](https://web.gencat.cat/)** - For funding and the RIS3-MCAT platform
- **[AINA Project](https://projecteaina.cat/)** - For the AINA Challenge 2024 framework

## License

This model is released under the [Apache 2.0 License](https://www.apache.org/licenses/LICENSE-2.0), consistent with the base Salamandra model.

## Links

- **Training Dataset**: [SIRIS-Lab/impuls-query-parsing](https://huggingface.co/datasets/SIRIS-Lab/impuls-query-parsing)
- **Project Repository**: [github.com/sirisacademic/aina-impulse](https://github.com/sirisacademic/aina-impulse)
- **Base Model**: [BSC-LT/salamandra-7b-instruct-tools](https://huggingface.co/BSC-LT/salamandra-7b-instruct-tools)
- **AINA Project**: [projecteaina.cat](https://projecteaina.cat/)
- **SIRIS Academic**: [sirisacademic.com](https://sirisacademic.com/)