Rethinking stance detection: A theoretically-informed research agenda for user-level inference using language models

2025-02-04

AI Summary

Current stance detection research suffers from two key limitations: (1) a lack of grounding in psychological or related theoretical frameworks, and (2) predominant focus on message-level rather than user-level modeling. This paper introduces the first theory-driven, user-centric stance detection research agenda, reconceptualizing stance as a computable individual psychological attribute and advocating large language models (LLMs) as a unified framework for inference and integration. Methodologically, we integrate pretrained language models with psychological and pragmatic theories to construct a user-attribute-enhanced stance modeling architecture. Our core contributions are: (1) a theoretically grounded, user-centered definition of stance; (2) four actionable research pathways; and (3) a novel paradigm for stance detection that ensures theoretical rigor, empirical scalability, and social inclusivity.

Technology Category

• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Humans and AI: Intelligent User Interfaces

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Sentiment analysis and opinion mining

Abstract

Stance detection has emerged as a popular task in natural language processing research, enabled largely by the abundance of target-specific social media data. While there has been considerable research on the development of stance detection models, datasets, and application, we highlight important gaps pertaining to (i) a lack of theoretical conceptualization of stance, and (ii) the treatment of stance at an individual- or user-level, as opposed to message-level. In this paper, we first review the interdisciplinary origins of stance as an individual-level construct to highlight relevant attributes (e.g., psychological features) that might be useful to incorporate in stance detection models. Further, we argue that recent pre-trained and large language models (LLMs) might offer a way to flexibly infer such user-level attributes and/or incorporate them in modelling stance. To better illustrate this, we briefly review and synthesize the emerging corpus of studies on using LLMs for inferring stance, and specifically on incorporating user attributes in such tasks. We conclude by proposing a four-point agenda for pursuing stance detection research that is theoretically informed, inclusive, and practically impactful.

Problem

• Enhance stance detection with theoretical frameworks
• Focus on user-level stance in language models
• Incorporate psychological features in stance modeling

Innovation

• User-level stance detection
• Incorporating psychological features
• Leveraging pre-trained language models