---
license: apache-2.0
datasets:
- QCRI/AZERG-Dataset
language:
- en
base_model:
- mistralai/Mistral-7B-Instruct-v0.3
tags:
- STIX standard
- threat intelligence
- MITRE ATT&CK
---

# QCRI/AZERG-T3-Mistral
This model is a fine-tuned version of `mistralai/Mistral-7B-Instruct-v0.3` specialized for `Task 3: Related Pair Detection`. It has been trained on the [QCRI/AZERG-Dataset](https://huggingface.co/datasets/QCRI/AZERG-Dataset) to  determine if a semantic relationship exists between two STIX entities within a given context.

This is a specialist model designed for high performance on entity detection within the [AZERG framework](https://github.com/QCRI/azerg/).

## Intended Use
Use this model to extract potential STIX entities from a given security text passage.

## Example Prompt:

```
Instruction:
You are a helpful threat intelligence analyst. Your task is to identify if the source entity and the target entity in the provided text passage are semantically related. To help you, we provide all the possible relationship labels between the source and target entities. If any label applies to the relationship, answer YES. Otherwise, answer NO. Answer in the following format: <related>YES or NO</related>

Input:
- Source Entity: [SOURCE ENTITY (ENTITY TYPE)]
- Target Entity: [TARGET ENTITY (ENTITY TYPE)]
- Possible Relationship Labels: [STIX RELATIONSHIP LABELS]
- Text Passage: [INPUT TEXT]

Response:
```

## Citation
If you use this model, please cite our paper:

```
@article{lekssays2025azerg,
  title={From Text to Actionable Intelligence: Automating STIX Entity and Relationship Extraction},
  author={Lekssays, Ahmed and Sencar, Husrev Taha and Yu, Ting},
  journal={arXiv preprint arXiv:2507.16576},
  year={2025}
}
```