A graph-based multi-modal LLM flow for EDA tasks.
CUDA Version: 12.4
PyTorch Version: 1.16.0
# Install Rust
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
conda create -n vlsillm python=3.10
conda activate vlsillm
pip install torch torchvision torchaudio
python -m pip install -r requirements.txt
data_collection/ contains the BRIDGES data generation flow.
The dataset has been temporarily removed from Hugging Face due to a pending license review with the EDA tool provider.
BRIDGES data can be found here.
If you use BRIDGES, please cite the following paper:
@misc{li2025bridges,
title = {BRIDGES: Bridging Graph Modality and Large Language Models within EDA Tasks},
author = {Wei Li and Yang Zou and Christopher Ellis and Ruben Purdy and Shawn Blanton and José M. F. Moura},
year = {2025},
eprint = {2504.05180},
archivePrefix= {arXiv},
primaryClass = {cs.LG},
url = {https://arxiv.org/abs/2504.05180},
doi = {10.48550/arXiv.2504.05180}
}