This is the official implementation for the paper Benchmarking and Improving Compositional Generalization of Multi-aspect Controllable Text Generation which has been accepted to appear at the main conference of ACL 2024. If you have any questions, please feel free to create an issue or contact the email: ztq602656097@mail.ustc.edu.cn, lizhaoyi777@mail.ustc.edu.cn.
- About the dataset in compmctg benchmark, please check data.
- About the meta-mctg framework, please check meta-mctg.
- About the evaluation system, please check evaluation.
- About the construction of protocols in compmctg benchmark, please check compmctg_protocols.
@misc{zhong2024benchmarking,
title={Benchmarking and Improving Compositional Generalization of Multi-aspect Controllable Text Generation},
author={Tianqi Zhong and Zhaoyi Li and Quan Wang and Linqi Song and Ying Wei and Defu Lian and Zhendong Mao},
year={2024},
eprint={2404.04232},
archivePrefix={arXiv},
primaryClass={cs.CL}
}