---
license: cc-by-nc-4.0
datasets:
- passing2961/stark-summary
language:
- en
base_model:
- meta-llama/Llama-3.2-3B-Instruct
tags:
- conversational ai
- conversation summarization
---

# Ultron-Summarizer-3B Model Card

[🏠 Homepage](https://stark-dataset.github.io/) | [💻 Github](https://github.com/passing2961/Stark) | [📄 Arxiv](https://arxiv.org/abs/2407.03958) | [📕 PDF](https://arxiv.org/pdf/2407.03958)

## List of Provided Model Series
- **Ultron-Summarizer-Series:** [🤖 Ultron-Summarizer-1B](https://huggingface.co/passing2961/Ultron-Summarizer-1B) | [🤖 Ultron-Summarizer-3B](https://huggingface.co/passing2961/Ultron-Summarizer-3B) | [🤖 Ultron-Summarizer-8B](https://huggingface.co/passing2961/Ultron-Summarizer-8B)
- **Ultron 7B**: [🤖 Ultron-7B](https://huggingface.co/passing2961/Ultron-7B)

> 🚨 Disclaimer: All models and datasets are intended for research purposes only.

## Model Description
- **Repository:** [Code](https://github.com/passing2961/Stark)
- **Paper:** [Stark: Social Long-Term Multi-Modal Conversation with Persona Commonsense Knowledge](https://arxiv.org/abs/2407.03958)
- **Point of Contact:** [Young-Jun Lee](mailto:yj2961@kaist.ac.kr)

## Model Details
- **Model**: Ultron-Summarizer-3B is a fully open-source conversational summarizer that generates summaries for long-term conversations, including those with image-sharing turns.
- **Date**: Ultron-Summarizer-3B was trained in 2024.
- **Training Dataset**: [Stark-Summary](https://huggingface.co/datasets/passing2961/stark-summary)
- **Architecture**: Ultron-Summarizer-3B was trained on top of [LLaMA-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct).

## How to Use

## License and Recommendations

🚨 Ultron-Summarizer-3B is intended to be used for research purposes only. 

## Acknowledgement

This work was supported by a grant of the KAIST-KT joint research project through AI Tech Lab, Institute of convergence Technology, funded by KT [Project No. G01230605, Development of Task-oriented Persona-based Dialogue Generation Combining Multi-modal Interaction and Knowledge Modeling].

## Citation

If you find the resources in this repository useful, please cite our work:

```
@article{lee2024stark,
  title={Stark: Social Long-Term Multi-Modal Conversation with Persona Commonsense Knowledge},
  author={Lee, Young-Jun and Lee, Dokyong and Youn, Junyoung and Oh, Kyeongjin and Ko, Byungsoo and Hyeon, Jonghwan and Choi, Ho-Jin},
  journal={arXiv preprint arXiv:2407.03958},
  year={2024}
}
```