الملخص
Simulating individual behavior requires large language models (LLMs) to preserve persona traits while adapting to dynamic social contexts. However, general-purpose LLMs often flatten distinct personas, while task-specific tuning suffers from fragmentation and generalization. To overcome these challenges, we organize individual simulation into the \textbf{FONTS Taxonomy}, comprising five complementary capability dimensions: \emph{persona fidelity} (\textbf{F}), \emph{outcome realization} (\textbf{O}), \emph{behavioral naturalness} (\textbf{N}), \emph{trajectory coherence} (\textbf{T}), and \emph{social grounding} (\textbf{S}). Grounded in this taxonomy, we curate a standardized training corpus library of approximately 10 million instances across 14 representative datasets and present \textbf{Socio-Foundation}. Socio-Foundation decouples specialization from integration via a three-stage pipeline: learning task experts via DAPO, consolidating them into capability experts via off-policy distillation, and unifying them via multi-teacher on-policy distillation (MOPD). We also establish \textbf{IndiEval}, consolidating 29 metrics across the FONTS dimensions. Experiments show that Socio-Foundation outperforms its \textit{Qwen3-8B} base by 11.0 points and approaches frontier models such as \textit{GLM-5.2}, with ablations and out-of-distribution evaluations further demonstrating the effectiveness and generalization of our model.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Wang, L., Xie, W., Mou, X., Luo, Y., & Wei, Z. (2026). Socio-Foundation: A Model for Generalizable Individual Behavior Simulation via Hierarchical Capability Distillation. https://omanscience.com/ar/articles/socio-foundation-a-model-for-generalizable-individual-behavior-simulation-via-hierarchical-capability-distillation
MLA 9
Wang, Liang, et al. "Socio-Foundation: A Model for Generalizable Individual Behavior Simulation via Hierarchical Capability Distillation." https://omanscience.com/ar/articles/socio-foundation-a-model-for-generalizable-individual-behavior-simulation-via-hierarchical-capability-distillation.
شيكاغو (المؤلف–التاريخ)
Wang, Liang, Wenxuan Xie, Xinyi Mou, Yixin Luo, and Zhongyu Wei. 2026. "Socio-Foundation: A Model for Generalizable Individual Behavior Simulation via Hierarchical Capability Distillation." https://omanscience.com/ar/articles/socio-foundation-a-model-for-generalizable-individual-behavior-simulation-via-hierarchical-capability-distillation.
هارفارد
Wang, L., Xie, W., Mou, X., Luo, Y. and Wei, Z. (2026) 'Socio-Foundation: A Model for Generalizable Individual Behavior Simulation via Hierarchical Capability Distillation', Available at: https://omanscience.com/ar/articles/socio-foundation-a-model-for-generalizable-individual-behavior-simulation-via-hierarchical-capability-distillation.
فانكوفر
Wang L, Xie W, Mou X, Luo Y, Wei Z. Socio-Foundation: A Model for Generalizable Individual Behavior Simulation via Hierarchical Capability Distillation. https://omanscience.com/ar/articles/socio-foundation-a-model-for-generalizable-individual-behavior-simulation-via-hierarchical-capability-distillation
IEEE
L. Wang, W. Xie, X. Mou, Y. Luo, and Z. Wei, "Socio-Foundation: A Model for Generalizable Individual Behavior Simulation via Hierarchical Capability Distillation," https://omanscience.com/ar/articles/socio-foundation-a-model-for-generalizable-individual-behavior-simulation-via-hierarchical-capability-distillation.