arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2204.04026cs.CL
收藏

用于计算论证的公平论证语言建模

Fair and Argumentative Language Modeling for Computational Argumentation

University of Mannheim(曼海姆大学);Bocconi University(博科尼大学)

Carolin Holtermann, Anne Lauscher, Simone Paolo Ponzetto

摘要:

虽然NLP的许多工作都集中在测量和减轻语义空间中的刻板偏见,但解决计算论证中偏见的研究仍处于起步阶段。在本文中,我们解决这个研究空白,并进行了深入的调查,偏见的争论语言模型。为此,我们引入ABBA,专门为论证量身定制的偏差测量的新资源。我们利用我们的资源来评估论证微调和去偏置的影响,发现在基于transformer的语言模型使用轻量级的适配器为基础的方法,更可持续和参数效率比完全微调的内在偏见。最后,我们分析了语言模型去偏对论证质量预测(计算论证的下游任务)性能的潜在影响。我们的研究结果表明,我们能够成功地和可持续地消除偏见,一般和争论的语言模型,同时保留(有时提高)模型性能的下游任务。我们在www.example.com上提供所有实验代码和数据。

英文摘要:

Although much work in NLP has focused on measuring and mitigating stereotypical bias in semantic spaces, research addressing bias in computational argumentation is still in its infancy. In this paper, we address this research gap and conduct a thorough investigation of bias in argumentative language models. To this end, we introduce ABBA, a novel resource for bias measurement specifically tailored to argumentation. We employ our resource to assess the effect of argumentative fine-tuning and debiasing on the intrinsic bias found in transformer-based language models using a lightweight adapter-based approach that is more sustainable and parameter-efficient than full fine-tuning. Finally, we analyze the potential impact of language model debiasing on the performance in argument quality prediction, a downstream task of computational argumentation. Our results show that we are able to successfully and sustainably remove bias in general and argumentative language models while preserving (and sometimes improving) model performance in downstream tasks. We make all experimental code and data available at https://github.com/umanlp/FairArgumentativeLM.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑