- Initial evaluation results are inside
MSC-evals
andTC-evals
directories, in json format. Edit or replace these files as per your need. These files were generated by the codes inside main repo. - Run eval_convert.py
- Run GraphEngine.ipynb
- Run PivotEngine.ipynb
If you find our work relevant and useful, please cite it in your work:
@misc{santra2023frugal,
title={Frugal Prompting for Dialog Models},
author={Bishal Santra and Sakya Basak and Abhinandan De and Manish Gupta and Pawan Goyal},
year={2023},
eprint={2305.14919},
archivePrefix={arXiv},
primaryClass={cs.CL}
}