Zero-Shot Stance Detection in the Wild: Dynamic Target Generation and Multi-Target Adaptation

TL;DR

This paper introduces a new zero-shot stance detection task in social media that automatically identifies multiple target-stance pairs without prior target knowledge, using large language models and novel evaluation metrics.

Contribution

It proposes the DGTA task, constructs a Chinese social media dataset, and explores fine-tuning strategies for LLMs to improve multi-target stance detection performance.

Findings

Fine-tuned LLMs outperform baseline models.

Two-stage fine-tuning achieves a target recognition score of 66.99%.

Integrated fine-tuning achieves an F1 score of 79.26%.

Abstract

Current stance detection research typically relies on predicting stance based on given targets and text. However, in real-world social media scenarios, targets are neither predefined nor static but rather complex and dynamic. To address this challenge, we propose a novel task: zero-shot stance detection in the wild with Dynamic Target Generation and Multi-Target Adaptation (DGTA), which aims to automatically identify multiple target-stance pairs from text without prior target knowledge. We construct a Chinese social media stance detection dataset and design multi-dimensional evaluation metrics. We explore both integrated and two-stage fine-tuning strategies for large language models (LLMs) and evaluate various baseline models. Experimental results demonstrate that fine-tuned LLMs achieve superior performance on this task: the two-stage fine-tuned Qwen2.5-7B attains the highest comprehensive target recognition score of 66.99%, while the integrated fine-tuned DeepSeek-R1-Distill-Qwen-7B achieves a stance detection F1 score of 79.26%.