README.md

---
license: mit
base_model:
- meta-llama/Meta-Llama-3-8B-Instruct
datasets:
- HuggingFaceH4/ultrachat_200k
- walledai/HarmBench
language:
- en
new_version: ASSELab/Diffusion-Llama-3-8B-Instruct
tags:
- pytorch
- llama
- llama-3
- DAT
- robust
- adversarial
library_name: transformers
---

# DAT - Distributional Adversarial Training

[![arXiv](https://img.shields.io/badge/arXiv-2511.04316-b31b1b.svg)](...)
[![GitHub](https://img.shields.io/badge/GitHub-DAT-181717?logo=github&logoColor=white)](https://github.com/ASSELab/DAT)

DAT utilizes [continuous adversarial training](https://arxiv.org/abs/2405.15589) on [diffusion-based](https://arxiv.org/abs/2511.00203v1) adversarial examples to close the gap between empirical and population-robust risk.
We fine-tune [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct).

#### This model is <u>**NOT**</u> using adversarial training! This is an ablation/baseline using just the diffusion data to fine-tune.

For further information, consult our paper []() or repository [https://github.com/ASSELab/DAT](https://github.com/ASSELab/DAT)

## Citation

```tex
@misc{,
      title={}, 
      author={},
      year={2026},
      eprint={},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}
```