FloatingDuck/zoom_model

1---2base_model: unsloth/llama-3.2-3b-instruct-bnb-4bit3tags:4- text-generation-inference5- transformers6- unsloth7- llama8- trl9license: cc10language:11- en12---13 14# Making Acoustic Side-Channel Attacks on Noisy Keyboards Viable with LLM-Assisted Spectrograms "Typo" Correction15 16## Model Overview17This is a fine-tuned version of the **LLaMA-3.2-3B** model for Acoustic Side-Channel Attacks (ASCA), designed to improve keystroke classification and error correction in noisy environments. The model leverages **Vision Transformers (VTs)** for spectrogram classification and **Large Language Models (LLMs)** for typo correction.18 19- **Fine-Tuned From**: [unsloth/llama-3.2-3b-instruct-bnb-4bit](https://huggingface.co/unsloth/llama-3.2-3b-instruct-bnb-4bit)20- **License**: CC21- **Developed by**: Seyyed Ali Ayati, Jin Hyun Park, Yichen Cai, Marcus Botacin22- **Repository**: [EchoCrypt GitHub](https://github.com/seyyedaliayati/EchoCrypt)23 24## Citation25 26If you use this model, please cite the following paper:27```bibtex28@article{ayati2025making,29  title={Making Acoustic Side-Channel Attacks on Noisy Keyboards Viable with LLM-Assisted Spectrograms' "Typo" Correction},30  author={Ayati, Seyyed Ali and Park, Jin Hyun and Cai, Yichen and Botacin, Marcus},31  journal={arXiv preprint arXiv:2504.11622},32  year={2025},33  url={https://arxiv.org/abs/2504.11622}34}35```36