Multi-label and Multi-target Sampling of Machine Annotation for Computational Stance Detection
This work addresses the problem of costly and time-consuming manual annotation for NLP researchers, offering an incremental improvement in automated labeling efficiency.
The paper tackles the challenge of scaling manual annotation for computational stance detection by investigating large language models for automated labeling, and introduces a multi-label and multi-target sampling strategy that significantly improves performance on benchmark corpora.
Data collection from manual labeling provides domain-specific and task-aligned supervision for data-driven approaches, and a critical mass of well-annotated resources is required to achieve reasonable performance in natural language processing tasks. However, manual annotations are often challenging to scale up in terms of time and budget, especially when domain knowledge, capturing subtle semantic features, and reasoning steps are needed. In this paper, we investigate the efficacy of leveraging large language models on automated labeling for computational stance detection. We empirically observe that while large language models show strong potential as an alternative to human annotators, their sensitivity to task-specific instructions and their intrinsic biases pose intriguing yet unique challenges in machine annotation. We introduce a multi-label and multi-target sampling strategy to optimize the annotation quality. Experimental results on the benchmark stance detection corpora show that our method can significantly improve performance and learning efficacy.