SCARLET

Official implementation of SCARLET: "Soft-Label Caching and Sharpening for Communication-Efficient Federated Distillation" (Accepted by IEEE TMC).

Important

The main branch contains a simplified implementation for better understanding of SCARLET’s core algorithms. For the exact experiment code and hyperparameter settings used in our paper, switch to the reproducibility branch.

Getting Started

uv

git clone https://github.com/kitsuyaazuma/SCARLET.git
cd SCARLET
uv sync

uv run python -m scarlet.main scarlet

Docker

docker run -it --rm --gpus=all --name scarlet ghcr.io/kitsuyaazuma/scarlet:main scarlet

# or

git clone https://github.com/kitsuyaazuma/SCARLET.git
cd SCARLET
docker build -t scarlet .
docker run -it --rm --gpus=all --name scarlet scarlet:latest scarlet

Configuration

All hyperparameters are managed with Tyro. You can see all available options by running:

uv run python -m scarlet.main --help

Citation

If you use this code in your research, please cite our preprint:

@ARTICLE{11344746,
  author={Azuma, Kitsuya and Nishio, Takayuki and Kitagawa, Yuichi and Nakano, Wakako and Tanimura, Takahito},
  journal={IEEE Transactions on Mobile Computing}, 
  title={Soft-Label Caching and Sharpening for Communication-Efficient Federated Distillation}, 
  year={2026},
  volume={},
  number={},
  pages={1-18},
  keywords={Servers;Computational modeling;Data models;Mobile computing;Entropy;Data privacy;Accuracy;Training;Quantization (signal);Federated learning;Federated learning;knowledge distillation;non-IID data;communication efficiency},
  doi={10.1109/TMC.2026.3652819}}