Zero-Shot Stance Detection using Contextual Data Generation with LLMs

Ghazaleh Mahmoudi, Babak Behkamkia, Sauleh Eetemadi

Track Selection: My submission fits into both tracks and I will let the PC decide if it is accepted.

Keywords: Stance Detection, ZeroShot, LLM

TL;DR: We generated a new dataset using VAST and GPT-3, in which each context is paired with more than one topic. Also We proposed a new pipeline for zero/few shot stance detection

Abstract: Stance detection, the classification of attitudes expressed in a text towards a specific topic, is vital for applications like fake news detection and opinion mining. However, the scarcity of labeled data remains a challenge for this task. To address this problem, we propose Dynamic Model Adaptation with Contextual Data Generation (DyMoAdapt) that combines FewShot Learning and Large Language Models. In this approach, we aim to fine-tune an existing model at test time. We achieve this by generating new topic-specific data using GPT-3. This method could enhance performance by allowing the adaptation of the model to new topics. However, the results did not increase as we expected. Furthermore, we introduce the Multi Generated Topic VAST (MGT-VAST) dataset, which extends VAST using GPT-3. In this dataset, each context is associated with multiple topics, allowing the model to understand the relationship between contexts and various potential topics.

Submission Number: 9