Skip-BART

Article: Zijian Zhao, Dian Jin, Zijing Zhou, Xiaoyu Zhang*, "Automatic Stage Lighting Control: Is it a Rule-Driven Process or Generative Task?" (under review)

Notice: We have uploaded the model, full parameters (RS2002/Skip-BART · Hugging Face), and dataset (RS2002/RPMC-L2 · Datasets at Hugging Face) in Hugging Face.

1. Model Structure

2. Dataset: RPMC-L2

The dataset is available at RPMC_L2. More details can be found at the paper.

3. How to Run

3. 1 Pre-train

python pretrain.py

3.2 Fine-tune

python finetune.py --model_path <pre-trained bart path>

4. Generate Your Own Stage Light

python generate.py --bart_path <fine-tuned backbone path> --head_path <fine-tuned model head path> --music_file <music file path>

The trained parameters are available at trained.zip .

5. Citation

@article{zhao2025automatic,
  title={Automatic Stage Lighting Control: Is it a Rule-Driven Process or Generative Task?},
  author={Zhao, Zijian and Jin, Dian and Zhou, Zijing and Zhang, Xiaoyu},
  journal={arXiv preprint arXiv:2506.01482},
  year={2025}
}