Modeling Epistemic Uncertainty in Social Perception via Rashomon Set Agents

2026-03-21

AI Summary

This study investigates the formation and evolution of students’ subjective perceptions of classroom social structures under conditions of incomplete information and individual perceptual differences. To this end, the authors propose a large language model–based multi-agent probabilistic framework that constructs personalized subjective graphs for each student, simulating their social perception, belief updating, and peer interactions within local informational constraints. The framework eschews global perspective assumptions by incorporating a Rashomon set, models perceptual biases through social anxiety–induced perturbations, and enables localized belief propagation via narrative exchanges with uncertainty-aware labels. Relying solely on local information, the model successfully replicates collective cognitive dynamics observed in real educational settings, with its predictions aligning closely with trends in actual student performance across six consecutive exam cycles.

Technology Category

• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Multiagent Systems: Adversarial Agents
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• User Modeling, Personalization and Recommendation: Social recommender systems and personalization for the social Web
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

We present an LLM-driven multi-agent probabilistic modeling framework that demonstrates how differences in students' subjective social perceptions arise and evolve in real-world classroom settings, under constraints from an observed social network and limited questionnaire data. When social information is incomplete and the accuracy of perception differs between students, they can form different views of the same group structure from local cues they can access. Repeated peer communication and belief updates can gradually change these views and, over time, lead to stable group-level differences. To avoid assuming a global "god's-eye view," we assign each student an individualized subjective graph that shows which social ties they can perceive and how far information is reachable from their perspective. All judgments and interactions are restricted to this subjective graph: agents use retrieval-augmented generation (RAG) to access only local information and then form evaluations of peers' competence and social standing. We also add structural perturbations related to social-anxiety to represent consistent individual differences in the accuracy of social perception. During peer exchanges, agents share narrative assessments of classmates' academic performance and social position with uncertainty tags, and update beliefs probabilistically using LLM-based trust scores. Using the time series of six real exam scores as an exogenous reference, we run multi-step simulations to examine how epistemic uncertainty spreads through local interactions. Experiments show that, without relying on global information, the framework reproduces several collective dynamics consistent with real-world educational settings. The code is released at https://anonymous.4open.science/r/Rashomonomon-0126.

Problem

• epistemic uncertainty
• social perception
• multi-agent modeling
• subjective graph
• Rashomon set

Innovation

• Rashomon set agents
• epistemic uncertainty
• subjective graph
• retrieval-augmented generation (RAG)
• LLM-driven multi-agent simulation