A novel benchmark to evaluate fine-grained individual role affection level and LLM's character relationship fidelity.
@inproceedings{kong2025sharp,
title={Sharp: Unlocking interactive hallucination via stance transfer in role-playing llms},
author={Kong, Chuyi and Luo, Ziyang and Lin, Hongzhan and Fan, Zhiyuan and Fan, Yaxin and Sun, Yuxi and Ma, Jing},
booktitle={Findings of the Association for Computational Linguistics: ACL 2025},
pages={839--866},
year={2025}
}… See the full description on the dataset page:
https://huggingface.co/datasets/furau/SHARP.
