Abstract:Stance detection automatically detects the stance in a text towards a target, vital for content analysis in web and social media research. Despite their promising capabilities, LLMs encounter challenges when directly applied to stance detection. First, stance detection demands multi-aspect knowledge, from deciphering event-related terminologies to understanding the expression styles in social media platforms. Second, stance detection requires advanced reasoning to infer authors' implicit viewpoints, as stance are often subtly embedded rather than overtly stated in the text. To address these challenges, we design a three-stage framework COLA (short for Collaborative rOle-infused LLM-based Agents) in which LLMs are designated distinct roles, creating a collaborative system where each role contributes uniquely. Initially, in the multidimensional text analysis stage, we configure the LLMs to act as a linguistic expert, a domain specialist, and a social media veteran to get a multifaceted analysis of texts, thus overcoming the first challenge. Next, in the reasoning-enhanced debating stage, for each potential stance, we designate a specific LLM-based agent to advocate for it, guiding the LLM to detect logical connections between text features and stance, tackling the second challenge. Finally, in the stance conclusion stage, a final decision maker agent consolidates prior insights to determine the stance. Our approach avoids extra annotated data and model training and is highly usable. We achieve state-of-the-art performance across multiple datasets. Ablation studies validate the effectiveness of each design role in handling stance detection. Further experiments have demonstrated the explainability and the versatility of our approach. Our approach excels in usability, accuracy, effectiveness, explainability and versatility, highlighting its value.

Leveraging Target-Oriented Information for Stance Classification

Stance Classification with Target-specific Neural Attention

Stance Detection with Target and Target Towards Attention

Improved Target-specific Stance Detection on Social Media Platforms by Delving into Conversation Threads

Guiding Computational Stance Detection with Expanded Stance Triangle Framework

Neural Multi-Task Learning for Stance Prediction

A Semi-Supervised Approach to Detecting Stance in Tweets

Multi-modal Stance Detection: New Datasets and Model

Enhanced Memory Network for Stance Classification

Integrating N-Gram Features into Pre-trained Model: A Novel Ensemble Model for Multi-target Stance Detection

Acquired TASTE: Multimodal Stance Detection with Textual and Structural Embeddings

Commonsense Knowledge Enhanced Memory Network for Stance Classification

How would Stance Detection Techniques Evolve after the Launch of ChatGPT?

Your Stance is Exposed! Analysing Possible Factors for Stance Detection on Social Media

Implicit Stance Detection with Hashtag Semantic Enrichment

Stance Detection with Collaborative Role-Infused LLM-Based Agents

Chain of Stance: Stance Detection with Large Language Models

Robust Stance Detection: Understanding Public Perceptions in Social Media

Cross-Target Stance Detection: A Survey of Techniques, Datasets, and Challenges

Investigating Chain-of-thought with ChatGPT for Stance Detection on Social Media

Incorporate opinion-towards for stance detection