Social Processes: Probabilistic Meta-learning for Adaptive Multiparty Interaction Forecasting

2025-01-03

AI Summary

This paper addresses the cross-group generalization challenge in multi-person social activities (e.g., conversational groups), arising from individual behavioral adaptability and inter-group interaction heterogeneity. We propose the first meta-learning framework for group-level dynamic social behavior prediction. Methodologically, each group is treated as an independent task, and we introduce the Social Process model: it jointly predicts multimodal future cue distributions for all group members while explicitly encoding historical interaction dynamics, integrating probabilistic meta-learning, multimodal temporal modeling, and variational inference. A novel contribution is the incorporation of synthetic-data-driven interpretability analysis. Evaluated on a photorealistic synthetic dataset, our approach achieves significant improvements in predictive accuracy for unseen groups and enhances semantic consistency in the learned latent space.

Technology Category

• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Multimodal Learning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• 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

Adaptively forecasting human behavior in social settings is an important step toward achieving Artificial General Intelligence. Most existing research in social forecasting has focused either on unfocused interactions, such as pedestrian trajectory prediction, or on monadic and dyadic behavior forecasting. In contrast, social psychology emphasizes the importance of group interactions for understanding complex social dynamics. This creates a gap that we address in this paper: forecasting social interactions at the group (conversation) level. Additionally, it is important for a forecasting model to be able to adapt to groups unseen at train time, as even the same individual behaves differently across different groups. This highlights the need for a forecasting model to explicitly account for each group's unique dynamics. To achieve this, we adopt a meta-learning approach to human behavior forecasting, treating every group as a separate meta-learning task. As a result, our method conditions its predictions on the specific behaviors within the group, leading to generalization to unseen groups. Specifically, we introduce Social Process (SP) models, which predict a distribution over future multimodal cues jointly for all group members based on their preceding low-level multimodal cues, while incorporating other past sequences of the same group's interactions. In this work we also analyze the generalization capabilities of SP models in both their outputs and latent spaces through the use of realistic synthetic datasets.

Problem

• Social Interaction Prediction
• Adaptability of Individual Behavior
• Inter-group Dynamics

Innovation

• Multi-person Interaction Prediction
• Social Process Model
• Group-specific Learning