🍻2024-11-29: Accepted to COLING 2025
🗒️ arXiv link: https://arxiv.org/abs/2409.07123
Cross-Refine has the following pipeline as shown in the figure:
- The
generator
outputs an initial explanation based on the given input (experiments/generator_explanation.py
)- The explanation will be judged by LLMs whether it needs to be improved further (
experiments/generator_explanation_evaluation.py
)
- The explanation will be judged by LLMs whether it needs to be improved further (
- The
critic
takes the initial explanation and input into account and outputs feedback on the explanation generated bygenerator
& suggested explanation (critic_feedback_explanation_generation.py
) - The
generator
gets feedback and suggested explanation fromcritic
and use them to refine the initial explanation (feedback_refinement.py
)
python -m pip install --upgrade pip
pip install -r requirements.txt
Here is an example for an instance from ECQA dataset:
@misc{wang2024crossrefineimprovingnaturallanguage,
title={Cross-Refine: Improving Natural Language Explanation Generation by Learning in Tandem},
author={Qianli Wang and Tatiana Anikina and Nils Feldhus and Simon Ostermann and Sebastian Möller and Vera Schmitt},
year={2024},
eprint={2409.07123},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2409.07123},
}