arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.14101cs.CL
收藏

通过少样本标注者适应实现高成本效益的主观任务标注与建模

Cost-Efficient Subjective Task Annotation and Modeling through Few-Shot Annotator Adaptation

University of Southern California(南加州大学)

Preni Golazizian, Alireza S. Ziabari, Ali Omrani, Morteza Dehghani

AI总结:

针对主观 NLP 任务标注预算受限问题,提出两阶段少样本标注者适应框架,并发布 24 人标注的道德情感语料,仅用 25% 预算即超越 SOTA 且提升公平性。

摘要:

在不存在单一标准答案的主观 NLP 任务中,纳入多样化标注者至关重要,因为他们的独特视角会显著影响标注结果。在现实场景中,标注预算往往成为决定数据及后续建模所纳入视角数量(即标注者数量)的主要因素。我们提出了一种用于主观任务标注收集与建模的新颖框架,旨在最小化标注预算,同时最大化每位标注者的预测性能。该框架采用两阶段设计:首先,我们依靠一小部分标注者构建一个多任务模型;其次,我们通过为每位标注者策略性地标注少量样本,为新视角增强该模型。为大规模测试该框架,我们引入并发布了一个独特数据集 Moral Foundations Subjective Corpus,其中包含由 24 位标注者标注道德情感的 2000 条 Reddit 帖子。我们证明,在两个数据集上,本框架仅使用原标注预算的 25%,就能在捕捉标注者个人视角方面超越此前的 SOTA。此外,本框架产生了更公平的模型,缩小了标注者之间的性能差距。

英文摘要:

In subjective NLP tasks, where a single ground truth does not exist, the inclusion of diverse annotators becomes crucial as their unique perspectives significantly influence the annotations. In realistic scenarios, the annotation budget often becomes the main determinant of the number of perspectives (i.e., annotators) included in the data and subsequent modeling. We introduce a novel framework for annotation collection and modeling in subjective tasks that aims to minimize the annotation budget while maximizing the predictive performance for each annotator. Our framework has a two-stage design: first, we rely on a small set of annotators to build a multitask model, and second, we augment the model for a new perspective by strategically annotating a few samples per annotator. To test our framework at scale, we introduce and release a unique dataset, Moral Foundations Subjective Corpus, of 2000 Reddit posts annotated by 24 annotators for moral sentiment. We demonstrate that our framework surpasses the previous SOTA in capturing the annotators' individual perspectives with as little as 25% of the original annotation budget on two datasets. Furthermore, our framework results in more equitable models, reducing the performance disparity among annotators.

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。

↑