Skip to content

[NeurIPS '24] On the Noise Robustness of In-Context Learning for Text Generation

License

Notifications You must be signed in to change notification settings

ml-stat-Sustech/Local-Perplexity-Ranking

Repository files navigation

On the Noise Robustness of In-Context Learning for Text Generation

This repository is the official implementation of On the Noise Robustness of In-Context Learning for Text Generation at NeurIPS 2024

Usage

We use Python 3.9.19, and other packages can be installed by:

pip install -r requirements.txt

Selecting demonstrations and generating output:

python main.py

Citation

If you find this useful in your research, please consider citing:

@article{gao2024noise,
  title={On the Noise Robustness of In-Context Learning for Text Generation},
  author={Gao, Hongfu and Zhang, Feipeng and Jiang, Wenyu and Shu, Jun and Zheng, Feng and Wei, Hongxin},
  booktitle={Advances in Neural Information Processing Systems (NeurIPS)},
  year={2024}
}

Acknowledgements

Our code is inspired by OpenICL. We thank the authors for releasing their code.

About

[NeurIPS '24] On the Noise Robustness of In-Context Learning for Text Generation

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages