ncarraz/prompt_generalization

ICLR 2023: Can discrete information extraction prompts generalize across language models?

Prompt generalization

This repository gathers the data and code for the paper Can discrete information extraction prompts generalize across language models?

We share all the generated prompts in this repository. Please refer to the corresponding fork if you want to evaluate the prompts on the language models or generate your own.

• Evaluation: LAMA
• Single-model training: AutoPrompt
• Mixed-training: Mixed-training AutoPrompt

Citation

@inproceedings{rakotonirina2023,
    author       = {Nathana{\"{e}}l Carraz Rakotonirina and
                    Roberto Dess{\`{\i}} and
                    Fabio Petroni and
                    Sebastian Riedel and
                    Marco Baroni},
    title        = {Can discrete information extraction prompts generalize across language
                    models?},
    booktitle    = {The Eleventh International Conference on Learning Representations,
                    {ICLR} 2023, Kigali, Rwanda, May 1-5, 2023},
    publisher    = {OpenReview.net},
    year         = {2023},
    url          = {https://openreview.net/pdf?id=sbWVtxq8-zE},
  }