Abstract
"传统立场检测通常假设目标已知,且仅输出立场类别(支持,反对,中立),难以应对目标不确定、立场判断需要有具体依据的情形。为此,本文提出目标自适应的可解释立场检测新任务,定义模型的输出为目标、观点和立场标签。具体地,构建了首个中文高质量立场检测数据集,并设计多维评估标准;评估了多种大语言模型的基线性能。实验发现:DeepSeek-V3在目标识别与立场分类表现最优,GPT-4o在观点生成上领先;大语言模型在目标明确时具备较强目标自适应能力,但处理存在反讽现象的输入时性能下降。数据集和实验结果公布于https://github.com/Cassieyy1102/TAISD。"
- Anthology ID:
- 2025.ccl-1.23
- Volume:
- Proceedings of the 24th China National Conference on Computational Linguistics (CCL 2025)
- Month:
- August
- Year:
- 2025
- Address:
- Jinan, China
- Editors:
- Maosong Sun, Peiyong Duan, Zhiyuan Liu, Ruifeng Xu, Weiwei Sun
- Venue:
- CCL
- SIG:
- Publisher:
- Chinese Information Processing Society of China
- Note:
- Pages:
- 298–310
- Language:
- URL:
- https://aclanthology.org/2025.ccl-1.23/
- DOI:
- Bibkey:
- Cite (ACL):
- Yi Lan, 王子豪 王子豪, Bo Chen, and Xiaobing Zhao. 2025. 目标自适应的可解释立场检测:新任务及大模型实验. In Proceedings of the 24th China National Conference on Computational Linguistics (CCL 2025), pages 298–310, Jinan, China. Chinese Information Processing Society of China.
- Cite (Informal):
- 目标自适应的可解释立场检测:新任务及大模型实验 (Lan et al., CCL 2025)
- Copy Citation:
- PDF:
- https://aclanthology.org/2025.ccl-1.23.pdf