Towards Characterizing Subjectivity of Individuals through Modeling Value Conflicts and Trade-offs

2025-04-17

AI Summary

This work addresses the challenge of modeling individual subjectivity—particularly for identifying and quantifying value conflicts and trade-offs in social media contexts to infer moral judgments. To this end, we propose SOLAR, the first framework to systematically model subjectivity at the individual level. SOLAR grounds its approach in value abstraction as a cognitive representation, integrating LLM-driven value conflict detection, text-based value trade-off modeling, and an interpretable subjective reasoning mechanism. Compared to prior methods, SOLAR significantly improves accuracy in moral judgment inference—especially in contentious scenarios—while enabling fine-grained, attribution-aware explanations of value preferences. This enhances model transparency and trustworthiness. The framework establishes a novel paradigm for subjectivity modeling in computational social science and trustworthy AI, advancing both theoretical understanding and practical deployment of value-sensitive systems.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Knowledge Representation and Reasoning: Preferences
• Philosophy and Ethics of AI: Accountability, Interpretability & Explainability

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Large Language Models (LLMs) not only have solved complex reasoning problems but also exhibit remarkable performance in tasks that require subjective decision making. Existing studies suggest that LLM generations can be subjectively grounded to some extent, yet exploring whether LLMs can account for individual-level subjectivity has not been sufficiently studied. In this paper, we characterize subjectivity of individuals on social media and infer their moral judgments using LLMs. We propose a framework, SOLAR (Subjective Ground with Value Abstraction), that observes value conflicts and trade-offs in the user-generated texts to better represent subjective ground of individuals. Empirical results show that our framework improves overall inference results as well as performance on controversial situations. Additionally, we qualitatively show that SOLAR provides explanations about individuals' value preferences, which can further account for their judgments.

Problem

• Characterize individual subjectivity via value conflicts modeling
• Infer moral judgments from social media using LLMs
• Improve inference with value preference explanations

Innovation

• LLMs model individual subjectivity via value conflicts
• SOLAR framework abstracts values from user texts
• Explains value preferences behind moral judgments